EUROPEAN COMMISSION DG Competition

Only the English text is available and authentic.

REGULATION (EC) No 139/2004 MERGER PROCEDURE

Article 6(1)(b) NON-OPPOSITION Date: 27/06/2016

In electronic form on the EUR-Lex website under document number 32016M7902

EUROPEAN COMMISSION

Brussels, 27.6.2016 C(2016) 4097 final

In the published version of this decision, some information has been omitted pursuant to Article 17(2) of Council Regulation (EC) No 139/2004 concerning non-disclosure of business secrets and other confidential information. The omissions are shown thus […]. Where possible the information omitted has been replaced by ranges of figures or a general description.

PUBLIC VERSION

MERGER PROCEDURE

To the notifying party:

Dear Sir/Madam,

Subject: Case M.7902 Marriott/Starwood Commission decision pursuant to Article 6(1)(b) of Council Regulation 1 No 139/2004and Article 57 of the Agreement on the European Economic 2Area

1OJ L 24, 29.1.2004, p. 1 ("the Merger Regulation"). With effect from 1 December 2009, the Treaty on the Functioning of the European Union ('TFEU') has introduced certain changes, such as the replacement of 'Community' by 'Union' and 'common market' by 'internal market'. The terminology of the TFEU will be used throughout this decision.

2OJ L 1, 3.1.1994, p.3 ("the EEA Agreement").

Commission européenne, DG COMP MERGER REGISTRY, 1049 Bruxelles, BELGIQUE Europese Commissie, DG COMP MERGER REGISTRY, 1049 Brussel, BELGIË

Tel: +32 229-91111. Fax: +32 229-64301. E-mail: COMP-MERGER-REGISTRY@ec.europa.eu .

1 THE PARTIES ............................................................................................................ 3

2 THE OPERATION...................................................................................................... 4

3 EU DIMENSION ........................................................................................................ 4

4 MARKET DEFINITION ............................................................................................ 4

4.1 Introduction ....................................................................................................... 4

4.2 Relevant product markets .................................................................................. 6

4.2.1 Hotel accommodation services ............................................................ 6

4.2.2 Hotel management services ............................................................... 20

4.2.3 Hotel franchising services ................................................................. 26

4.3 Relevant geographic markets .......................................................................... 27

4.3.1 Hotel accommodation services .......................................................... 27

4.3.2 Hotel management services ............................................................... 31

4.3.3 Hotel franchising services ................................................................. 37

5 COMPETITIVE ASSESSMENT .............................................................................. 42

5.1 Introduction ..................................................................................................... 42

5.2 The markets for hotel accommodation services .............................................. 43

5.2.1 Methodology for the estimation of market shares on markets for hotel accommodation services ........................................................... 43

5.2.2 Introduction on affected markets ....................................................... 47

5.2.3 Elements common to the assessment of markets for hotel accommodation services in the EEA ................................................. 49

5.2.4 Assessment at local level ................................................................... 55

5.3 Hotel management services ............................................................................. 77

5.3.1 Methodology for estimating market shares ....................................... 77

5.3.2 Estimated market shares of the Parties and affected markets ............ 80

5.3.3 Effects of the Transaction on ongoing management agreements ...... 80

5.3.4 Effects of the Transaction on new management agreements ............ 81

5.3.5 Conclusion ......................................................................................... 87

5.4 Hotel franchising services ............................................................................... 87

5.4.1 Estimated market shares of the Parties and affected market ............. 87

5.4.2 Effects of the Transaction on ongoing franchise agreements ............ 88

5.4.3 Effects of the Transaction on new franchise agreements .................. 88

5.4.4 Conclusion ......................................................................................... 90

6 CONCLUSION ......................................................................................................... 90

(1) On 23 May 2016, the European Commission received a notification of a proposed concentration pursuant to Article 4 of Council Regulation (EC) No 139/2004 by which the undertaking Marriott International, Inc. ("Marriott") acquires within the meaning of Article 3(1)(b) of the Merger Regulation sole control of the undertaking Starwood Hotels & Resorts Worldwide, Inc. ("Starwood"). Marriott and Starwood will hereinafter collectively be referred to as "the Parties".

1 THE PARTIES

(2) Marriott is a diversified hospitality company which acts as a manager and franchisor (and on rare occasions, owner) of hotels and timeshare properties in 85 countries and territories. Marriott owns 19 hotel brands internationally, including: The Ritz-Carlton, EDITION, JW Marriott, Autograph Collection Hotels, Renaissance Hotels, Marriott Hotels, Delta Hotels and Resorts, Marriott Executive Apartments, Marriott Vacation Club, Gaylord Hotels, AC Hotels by Marriott, Courtyard, Residence Inn, SpringHill Suites, Fairfield Inn & Suites, TownePlace Suites, Protea Hotels and Moxy Hotels. In addition, it has a license agreement to utilize the Bulgari brand for hotels. As of 30 June 2016, there will be 275 hotels in the EEA operating under Marriott brands, with […] rooms.

(3) Marriott has a loyalty program marketed under two brands, Marriott Rewards and Ritz-Carlton Rewards.

(4) Starwood is also a manager and franchisor (and on rare occasions, owner or leaseholder) of hotels and resorts worldwide, with nearly 1 300 properties in some 100 countries. Starwood owns the following hotel brands: St. Regis, The Luxury Collection, W, Westin, Le Méridien, Sheraton, Four Points by Sheraton, Aloft, Element, and the recently introduced Tribute Portfolio. As of 30 June 2016, there will be 145 hotels in the EEA operating under Starwood brands, with […] rooms. Starwood owns and operates […] of these hotels […], with […] rooms […]. Starwood manages […] of these hotels […], with […] rooms […], under management agreements for third-party owners. […] of these hotels […], with […] rooms […], are franchised hotels, for which Starwood is neither the owner nor the operator of the hotel.

3 Publication in the Official Journal of the European Union No C 192, 31.05.2016, p. 4.

4 Marriott currently owns fewer than ten hotels of the more than 4 500 hotels in the worldwide Marriott system.

5 Form CO, section 2.2.B.

6 Starwood currently owns/leases [fewer than ten] hotels in the EEA, […].

7 Until recently, Starwood also owned Starwood Vacation Ownership, Inc., which is a timeshare business which has activities in the USA, Mexico and the US Virgin Islands. The Transaction is conditional on disposal of this business from Starwood. This condition has been fulfilled, as the timeshare business was sold to Interval Leisure Group, Inc. (“ILG”) on 11 May 2016.

8 Not including Design Hotels. Design Hotels is a marketing affiliation for a group of independent hotels, none of which is owned, managed, or franchised by Starwood. As such, these hotels have been treated as independent hotel competitors in the assessment of the Transaction.

9 Form CO, section 2.2.C.

(5) Starwood’s loyalty program is Starwood Preferred Guest.

2 THE OPERATION

(6) Pursuant to an agreement and plan of merger dated 15 November 2015 and amended on 20 March 2016, Marriott will acquire sole control over Starwood. The Transaction is a result of a series of internal preparatory steps, following which Starwood will be absorbed by a wholly-owned subsidiary of Marriott. As a result, Marriott will be the sole “parent” company of the Starwood group and therefore through the Transaction, Marriott would acquire sole control of Starwood. Marriott's Charter and Bylaws will govern the merged entity post-Transaction. All fourteen board members of the merged entity will be appointed by Marriott and have a fiduciary duty to Marriott's shareholders.

(7) Therefore, the Transaction constitutes a concentration within the meaning of Article 3(1)(b) of the Merger Regulation.

3 EU DIMENSION

(8) The undertakings concerned have a combined aggregate worldwide turnover of more than EUR 5 000 million (Marriott: EUR 22.6 billion, Starwood: EUR 11.8 billion). Each of them has a Union-wide turnover in excess of EUR 250 million (Marriott: EUR […], Starwood: EUR […]), but they do not achieve more than two-thirds of their aggregate Union-wide turnover within one and the same Member State.

(9) The concentration therefore has a Union dimension under Article 1(2) of the Merger Regulation.

4 MARKET DEFINITION

4.1 Introduction

(10) The Parties are both active in the hotel industry. More concretely, 275 hotels operate under Marriott brands in the EEA, with a total of […] rooms. […] hotels are operated under Starwood brands in the EEA, with a total of […] rooms.

(11) The operation of a hotel may involve up to three different parties:

12(a) the entity that owns or leases (as the tenant) the hotel;

(b) the entity that manages the hotel; and

10 Form CO, paragraph 50.

11 Turnover calculated in accordance with Article 5 of the Merger Regulation and the Commission Consolidated Jurisdictional Notice (OJ C 95, 16.4.2008, p. 1). For hotels, which are jointly controlled by owners and hotel operators, 50% of the turnover of the Parties’ managed hotels is included in the Parties’ turnover, in line with paragraph 186 of Commission’s Consolidated Jurisdictional Notice.

(c) the entity that owns the brand under which the hotel operates.

(12) As a result, hotels may be operated under one of the three following main business models:

(a) Under the first model, hotels are owned and managed by the same company under its own name or brand, which has sole control over them. Hotels operated under this model will be referenced as "owned or leased hotels" in this Decision.

(b) Under the second model, hotels are managed by a company − hotel chain or specialised management company ("white label management companies") − on behalf of their owner for a management fee. These hotels are operated under the name or brand of the manager, notably if the latter is a hotel chain, or the owner, if the manager is a white label management company. Hotels operated under this model will be referenced as "managed hotels" in this Decision.

(c) Under the third model, hotel chains franchise one of their brands to hotel owners, who either manage their hotels themselves or use a third management company. Hotels operated under this model will be referenced as "franchised hotels" in this Decision.

(13) Marriott submits that hotels operating under the Marriott and the Starwood brands belong to all three categories. More concretely, as of 30 June 2016, Marriott owns […] and leases […] hotels with a total of […] rooms in the EEA, manages […] hotels with a total of […] rooms on behalf of their owners and franchises its brands to another […] hotels with a total of […] rooms. Starwood owns […] hotels with […] rooms in the EEA, manages […] on behalf of their owners with a total of […] and franchises […] with […] rooms.

(14) According to Marriott, companies active in the sector may choose to operate under different models, some owning or leasing a greater proportion of hotels operating under their brands, whereas others may prefer managed or franchised hotels. Both Parties have over time adopted an asset-light business model, under which they separate ownership of the real estate from the hotel management business and the hotel franchising business and increasingly focus their activity on the latter two.

13 Form CO, paragraph 17.

14 Form CO, paragraph 28.

15 Form CO, paragraph 34.

16 Form CO, paragraph 19.

17 Form CO, paragraphs 26 and 32. As the Parties' activities in owning and leasing hotels are marginal and expected to further decrease, the Commission will not further consider them in this Decision. In particular, given the marginal scope of the Parties' activities in owning and leasing hotels, the Transaction is unlikely to give rise to any material vertical effect due to the Parties' activities in the provision of management and franchising services to hotel owners.

(15) This variety of business models and ways of operating a hotel has been confirmed by the Parties' competitors and other market participants. Indeed, several hotel chains, such as Hilton, Accor, IHG and others, indicate in the context of the market investigation that hotels operating under their brands may be owned/leased, managed or franchised and that, in addition to the operation of owned or leased hotels, they also engage in the provision of hotel management and hotel franchising services.

(16) In light of the above, in the EEA, the Parties' activities overlap on markets for the provision of hotel accommodation services, markets for the provision of hotel management services and markets for the provision of hotel franchising services.

4.2 Relevant product markets

4.2.1 Hotel accommodation services

(17) The Commission has in its prior decision practice considered a separate market for the provision of hotel accommodation services, leaving however the exact product market definition open.

(18) Marriott submits that there is a spectrum of differentiated offerings in the hotel industry and that the market is heterogeneous from both the demand and the supply side. While the main service provided is accommodation, the sleeping areas vary in terms of size, quality and range of furnishing, location of the room within the hotel etc. Similarly, the range and quality of ancillary services may differ, as some hotels offer 24-hour front desk services, operate restaurants and bars, spa facilities, gift shops, offer meeting spaces, etc. Given the many different services that may or may not be provided in addition to hotel accommodation and the different levels of quality in the provided services, hotels can be differentiated in a multitude of ways. Any hypothetical means of segmentation however, will, according to Marriott, necessarily be highly imperfect as the boundaries between any sub-segments of the total market for hotel accommodation will be blurred. Finally, Marriott believes that the exact market definition can ultimately be left open.

(19) The majority of respondents to the market investigation having expressed an opinion also indicate that a further segmentation of the market for hotel accommodation services is not warranted.

(20) Whether the market for hotel accommodation services should be further segmented will be analysed below. More concretely, it will be assessed whether the overall market should be sub-segmented by ownership type (Section 4.2.1.1) and/or by comfort/price level (Section 4.2.1.2). Moreover, it will be assessed whether a distinct product market should be considered for accommodation services in short-stay residences (Section 4.2.1.3), or for the provision of specific services, such as conferences and events services, hotel loyalty programs, etc. (Section 4.2.1.4).

18 See notably, Commission's market reconstruction, Section 5.3.1 below.

19 Form CO, paragraph 96 and following.

20 See responses to Q1a – Questionnaire to competitors, question 6, Q1b – Questionnaire to independent and franchised competitors, question 5, Q2 – Questionnaire to corporate customers, question 2, Q3 – Questionnaire to consumer associations, question 3, Q5 – Questionnaire to travel agencies and tour operators, question 5.

4.2.1.1 Distinction by ownership type

4.2.1.1.1 Commission's practice

(21) In its prior decision practice, the Commission has considered a sub-segmentation of the overall market for hotel accommodation by ownership type. More concretely, the Commission distinguished between three types of hotels, namely (i) economically and legally independent hotels; (ii) voluntary chains consisting of groups of independent hotels which carry out their marketing, promotion, purchasing etc. under one and the same hotel brand; and (iii) integrated chains which operate hotels directly through subsidiaries or indirectly by a franchise or management contract.

(22) The Commission identified the following elements as distinguishing the offer of hotel chains from that of independent hotels. From the supply side, hotel chains are organised on the basis of a network concept which meets service requirements that go beyond the purely local framework, are much more uniform from one hotel to another and more extensive (e.g. extended opening hours, central reservation system, restaurants, etc.). Moreover, hotel chains operate under a common hotel name and trade mark and have a common marketing strategy for all hotels of the chain, which enables them to raise awareness of their brand much more effectively and at a lower cost than independent hotels would. They also use own centralised reservation systems or have access to international reservation systems, such as Amadeus, Galileo, etc. Last, hotel chains pursue a policy of actively seeking customers, by approaching travel agents, tour operators, corporate customers etc. and offering them differentiated rates, promotions and additional services in order to increase their total sales across the chain. Also from the demand side, large customers such as travel agents, tour operators, corporate customers etc. prioritise hotel chains, with which they can enter into framework contracts setting out negotiated conditions for prices, terms of payment, commissions and discounts.

(23) The Commission further acknowledged that independent hotels increasingly organise themselves in voluntary chains. In doing so, they become substitutable to hotel chains and increase the offer available to corporate customers, tour operators and travel agencies.

(24) Ultimately however, the Commission has always left open whether the market for the provision of hotel accommodation services should be further segmented on the basis of ownership type.

21 See among others, Case IV/M.126 – Accor / Wagons-Lits, point U; M.1596 - Accor / Blackstone / Colony / Vivendi, recitals 16 and following; M.6058 – Bank of Scotland / Barclays Bank / Kew Green Hotels, recital 13.

22 Case IV/M.126 – Accor / Wagons-Lits, point U, M.1596 - Accor / Blackstone / Colony / Vivendi, recitals 16 and following.

23 Case M.1596 - Accor / Blackstone / Colony / Vivendi, recitals 19 and following.

4.2.1.1.2 Marriott's views

(25) Marriott submits that independent and chain hotels are substitutable and in competition with each other and that the elements on the basis of which the Commission differentiated between the two in its prior practice do not apply to the industry today.

(26) More concretely, independent and chain hotels are in the position of offering similar types of services and amenities. Moreover, not all hotel chains offer uniform services among their hotels. The increased operation of hotels under management and franchise contracts allows for a more independent operation of hotels belonging to the same chain. The proliferation of information technology, internet travel websites and reservation platforms have changed the industry and placed chain and independent hotels on an equal footing, as they both have immediate access to millions of customers. Similarly, customers are also empowered to shift their bookings quickly and efficiently away from any hotel operator that fails to meet their expectations on price or quality/facilities. Last, both chain and independent hotels actively compete for corporate customers and cooperate with travel agents and tour operators, as both chain and independent hotels offer promotional packages and special prices.

(27) Marriott further points out that operators of independent hotels have a particularly strong presence in the EEA, corresponding to approximately 70% of all hotels. Also, chain hotels are in competition with independent hotels, as evidenced by the fact that they benchmark themselves against both chain and independent hotels. Excluding independent hotels from the relevant market and from the competitive assessment would therefore in Marriott's view fail to reflect market reality.

4.2.1.1.3 Commission's assessment

Membership to a chain

(28) Hotel operators consider themselves in competition with both independent and chain hotels. Indeed, the Parties' hotels' competitive sets, i.e. the list of hotels against which Marriott and Starwood hotels benchmark themselves and the performance of which they monitor, consist of both chain and independent hotels. For instance, [examples of independent hotels included in the Parties' competitive sets in Milan] are included in the Parties' competitor sets in Milan, [examples of independent hotels included in the Parties' competitive sets in Barcelona] in Barcelona etc.

(29) Furthermore, the Parties' competitors having expressed an opinion in the Commission's market investigation indicate that whether a hotel belongs to a chain or is independent is not among the main drivers of customers' choice between hotels. Instead, features such as price, comfort level and customers' ratings are more important for the selection of a hotel within a given location. Similarly, customers, as well as travel agents and tour operators, also explain that they select hotels taking into account first elements such as price and comfort level, as well as corporate policy, in the case of corporate customers, and users' ratings for end consumers in general. Features such as the affiliation with a chain, the brand and the offering of loyalty programs on the other hand are not among customers' main selection criteria.

24 Form CO, paragraphs 120 and following.

25 The competitor sets of the Parties' hotels are extracted from their revenue optimisation tools and in some instances also annexed to their management or franchise contracts; Form CO, paragraph 865 and following and Annexes 8.1.a.1, 8.1.a.2 and 8.1.b.1. Specific analyses at city level will also be conducted in Section 5.2.4 below.

26 See responses to Q1a – Questionnaire to competitors, question 5; Q1b – Questionnaire to independent and franchised competitors, question 4.

(30) The EEA travel industry has evolved significantly in the last decades, most notably through technological developments, such as the broader use of information technology and online tools and sources, and through the increasing number of hotel chains.

(32) Second, a number of companies offer technology solutions to hotels that facilitate the distribution of their services, enable direct bookings by customers, help hotel operators to establish access to travel agents’ and tour operators’ platforms and metasearch engines. Further services supporting the operation of a hotel, such as business intelligence media, distribution strategy support, revenue management tools and other services are also available on the market and may be used by chain and independent hotels alike. Therefore, independent hotels have nowadays access to technology solutions that give them the possibility to provide easy customer interface and improve their overall performance.

(33) The majority of respondents having expressed an opinion in the market investigation submit that chain and independent hotels are rather interchangeable regarding their presence on different distribution channels, but not regarding their access to centralised reservation systems. Against the background of the overall development of the

27 See responses to Q2 – Questionnaire to corporate customers, question 2.2, Q3 – Questionnaire to consumer associations, question 2, Q5 – Questionnaire to travel agencies and tour operators, question 4.

28 Form CO, paragraphs 129-131 and 135.

(34) Third, the overall evolution of the industry has further facilitated the organisation of independent hotels in either voluntary or integrated chains in recent years.

(35) In the case of voluntary chains, independent hotels set up or enter into affiliations that operate under a common brand, enabling them to promote each other's services. Members of voluntary chains share centralised resources and a wide spectrum of tools and services, such as marketing and advertising support, sales and distribution services, access to global distribution and internet reservation systems, revenue management tools, interconnectivity with OTA platforms, training services, quality systems, loyalty programs, etc. Some of these voluntary chains have a large geographical footprint and account for several hundreds of members, as for example Leading Hotels of the World, Worldhotels, Preferred Hotels, etc.

(36) Membership to an integrated chain has also evolved in recent years, as business models based on hotel franchising are more broadly used, whereas in the past chain hotels were typically owned and operated by the chain owner. Conversely, today, the number of hotel chains operating in the EEA under an "asset light" model, whereby they focus on the provision of management and franchising services to hotel owners rather than the ownership of their hotels has significantly increased. The more widespread use of a hotel operation model based on franchising has thus provided owners and operators of independent hotels with the option to join a hotel chain, while retaining a significant degree of independence. This way, they have the possibility to profit from the hotel chain's brand awareness, systems and tools and know-how, while maintaining control over the management of their hotel. The degree of independence of franchised hotels will be further analysed in Section 5.2.1.2 below.

Price rates, services and amenities

(37) Overall, with regard to the price rates and the services and amenities offered, no particular differences are identified between chain and independent hotels. The majority of respondents having expressed an opinion in the market investigation submit that chain and independent hotels are interchangeable in terms of price.

(38) Among hotels of similar comfort/price level, offerings such as 24h front desk coverage, gym and pool facilities, restaurants and bars are provided by both chain and independent hotels. The majority of respondents having expressed an opinion in the market investigation also consider that the range of services and amenities offered by chain hotels is interchangeable with that offered by independent hotels.

(39) As to the uniformity of the services offered from one hotel to another, the majority of respondents having expressed an opinion in the market investigation submit that chain hotels offer more uniform products and services from one hotel to another than independent hotels do. Several other respondents, however, indicate that the level of uniformity depends on the specific chains and brands and that, in particular in the case of voluntary chains or chains of franchised hotels, the services offered may vary significantly between hotels belonging to the same chain. Indeed, in a number of hotel chains, a significant degree of differentiation in the appearance, style and service offerings of member hotels is maintained. In the case of high end integrated chains, such as Marriott's Autograph Collection hotels, this may be a strategic choice to appeal more to customers that value individuality. In voluntary chains, it is a result of the business model that provides for cooperation in some aspects of member hotels' activity, but not for the standardisation of their offerings. In addition, offerings' uniformity has overtime become significantly less relevant as the increased use of OTAs and metasearch engines enables customers not only to acquire immediate information as to the various hotels' service offerings, but in most cases to also see pictures and read users' reviews about the quality and availability of the various services.

(40) A service feature, typical of chain hotels that are organised in a network, is the set-up of loyalty programs, rewarding customers' multiple stays in hotels of the chain. In recent years, certain loyalty programs are also available to customers of independent hotels. This is primarily facilitated through the participation of independent hotels into loyalty programs offered by aggregators, such as OTAs, metasearch engines, etc. Customers of independent hotels that have opted to participate in such OTA loyalty programs, receive rewards related to car rental companies, hospitality providers, etc. if they use the respective OTA for their bookings. The majority of respondents having expressed an opinion in the market investigation however submit that, with regard to loyalty programs, the offer of chain hotels is not substitutable for that of independent hotels.

Access to the customer base

(41) Independent and chain hotels alike have access to the same customer base. The technological developments and overall evolution of the sector not only helped independent hotels to raise awareness of their brands among a larger part of the customer base through more distribution channels and OTAs. These developments and evolution further enabled them to compete for corporate customers, tour operators and travel agents, which in the past appeared to prioritise chain hotels for their bookings. Indeed, even though the majority of respondents having expressed an opinion in the market investigation indicate that chain hotels have more recognition and can more easily reach or be reached by customers than independent hotels, they also submit that customers, including corporate customers, purchase hotel accommodation services from both chain and independent hotels.

(42) First, corporate customers explain that location is one of the main criteria for selecting a hotel accommodation, as they are interested in ensuring the proximity of their staff to the working premises. The requests for proposals they solicit are therefore addressed to hotel operators in the particular area of their interest, irrespective of whether these belong to a chain or not. Ultimately, corporate customers select a number of hotels, with which they enter into a contractual relationship, on the basis of the proposals submitted by the hotel operators. Even though the majority of corporate customers having expressed an opinion in the market investigation indicate that chain hotels have an easier access to corporate customers than independent hotels, notably due to their dedicated sales departments and ability to centralise purchases for a greater number of locations, they unanimously submit that they purchase hotel accommodation services from chain and independent hotels.

(43) Second, large OTAs submit that not only they include both independent and chain hotels in their inventory, but that the majority of their inventory corresponds to independent hotels. Moreover, OTA users do not seem to limit their search of hotel accommodation provider to chain hotels, as Booking.com submits that users do not usually filter based on specific brands and Expedia that it does not offer their users any filter function allowing them to distinguish between chain and independent hotels. The majority of respondents having expressed an opinion in the market investigation further indicate that OTAs prioritise offerings irrespective of whether the hotels belong to chains or are independent.

(44) The majority of travel agents and tour operators having expressed an opinion in the market investigation also indicate that they propose to their customers hotel accommodation services from both chain and independent hotels alike and that they give the same visibility and priority in their offerings to chain and independent hotels. Moreover, independent hotels may also offer promotional packages to tour operators and travel agents, following the practice of chain hotels. The Parties submit and travel agents and tour operators having expressed an opinion in to the Commission's market investigation also indicate that in most chain hotels, the pricing strategy is decided at the level of the hotel and not centrally by the headquarters for the entire chain. Therefore, the position of independent hotels is not significantly different from that of chain hotels, when it comes to their ability to make promotions and offer rebates to tour operators and travel agents. The majority of travel agents and tour operators having expressed an opinion in the market investigation also indicate that whether a hotel operator offers special prices, discounts, promotional packages, related services such as car rental etc. usually depends on each operator and its contractual relationship with the travel intermediary, rather than on its membership to a chain.

(45) Third, independent hotels may also pursue a policy of actively seeking customers, as they have access to a number of distribution channels and tools that may further increase their reputation. Illustratively, some OTAs use priority listings systems, giving greater visibility to hotels that are prepared to pay for this service. The fact that independent hotels are in some instances displayed among the first results of searches in OTA's websites, indicates that they are not only able, but actually making use of this opportunity to increase their customer base. Some travel agents and tour operators having expressed an opinion in the market investigation state that they offer some form of priority listing for hotel operators willing to pay for such service. Among these travel intermediaries, the majority indicates that whether a hotel operator will choose to make use of such priority listing is unrelated to it being a chain or an independent hotel.

(46) In light of the above considerations, the consumers' need to rely on brand as a signal of quality and likely performance has reduced in recent years. Chain and independent hotels offer a similar range of services and amenities at similar prices and have access to broad distribution networks and tools alike enabling them to reach the largest part of the customer base, raise awareness of their service offerings and provide user-friendly reservation systems.

(47) Therefore, the Commission concludes that, for the purposes of this Decision, the market for hotel accommodation services should not be further segmented on the basis of ownership type, but rather comprise hotel accommodation services supplied by both chain and independent hotels.

(48) However, because of, inter alia, their more limited access to centralised reservation platforms and loyalty schemes, independent hotels may exert a somewhat lower competitive pressure on chain hotels than chain hotels do on each other. This element will be taken into account in the competitive assessment (Section 5.2), in particular when assessing if the Parties compete closely with each other on the markets for hotel accommodation services.

4.2.1.2 Distinction by comfort/price level

4.2.1.2.1 Commission's practice

(49) The Commission has in its prior decision practice considered a possible segmentation of the market for hotel accommodation by comfort/price level. This segmentation was notably based on the star-rating of each particular hotel, the star rating being indicative of the standard and facilities the customer may expect when selecting a specific hotel.

(50) In past cases, the Commission has considered narrower sub-segments for hotels having the same star rating, for instance 4-star hotels only, 5-star hotels only, etc., as well as broader sub-segments of the total market including hotels with successive star ratings, such as 4- and 5-star hotels combined, 3- and 4-star hotels combined, 2-, 3- and 4-star combined, etc. Broader categories were taken into account, as even though the star rating reflects the standard and facilities customers may expect from a specific hotel, hotels belonging to successive star categories may in many instances be substitutable from a consumer point of view in terms of price, location and services.

(51) Ultimately, however, the Commission has always left open whether and in which way the overall market for hotel accommodation services should be sub-segmented by comfort/price level.

4.2.1.2.2 Marriott's views

(52) Marriott submits that the high degree of differentiation among hotels does not lend to a clear segmentation of the market by comfort or price level. Rather, the provision of hotel accommodation services comprises a spectrum, in which hotels offering diverse service and quality levels compete with hotels both up and down the spectrum. According to Marriott, there is a high degree of substitutability among portions of the spectrum and this substitutability often spans multiple star categories. Although a consumer may be unlikely to view a 5-star hotel and a 1-star hotel as substitutes for one another, customers can and do consider hotels with different star-ratings as potential substitutes, particularly those with adjacent star ratings, for instance 5-star and 4-star hotels; 4-star and 3-star; etc.

(53) Marriott further submits that the segmentation by category used in the industry does not constitute a reliable proxy for market definition purposes, as hotels with similar locations, amenities and price points that clearly compete with each other are classified into different categories.

4.2.1.2.3 Commission's assessment

(54) Hotel classification systems by comfort/price level exist in most Member States at national and regional level. A great number of hotel associations have contributed to the development of such systems, either on their own initiative or in collaboration with public authorities. As a result of the differences in culture and geographical situations, there are also significant differences between the criteria and methodology followed by the systems used in the various EEA countries.

(55) The most well-known of these classification systems appear to be the star-rating system, which is widely used and most visible to consumers, and the class-

Segmentation by star rating

(56) The star-rating system provides generally for a rating of 1 to 5 stars, granted to hotels on the basis of certain criteria linked to the type of services they offer and their quality.

(57) Against the background of the various national or regional classification systems using diverse criteria, star-ratings may not always provide equivalent comparisons between hotels located in different geographic areas. Within a particular country or city, however, where a single star-rating system would be applied, the number of stars awarded to hotels would constitute a relevant proxy for comparing their offerings. Indeed, by comparing the services and amenities, as well as the price of hotels with the same star rating in the same location, greater similarities may be identified than when comparing them with hotels with a different star rating. Nevertheless, individual 4-star hotels for example may at times offer a broader service range or be priced higher than 5-star hotels in the same location.

(58) Several sophisticated customers and competitors having expressed an opinion in the market investigation indicate that in light of these discrepancies and the fact that customers nowadays rely more on users' ratings for their hotel selection, star rating systems are losing in significance. The majority of competitors having expressed an opinion in the market investigation thus consider that a segmentation of the market by star rating is not required.

(59) On the other hand, star rating appears to be a significant criterion for at least part of the customer base, as reflected in its widespread use by consumers throughout the world, notably through online travel agents. Among the customers having expressed an opinion in the market investigation that consider a segmentation of the total market by comfort/price level warranted, the majority submits that this segmentation should be done by star rating.

(60) The majority of respondents having expressed an opinion in the market investigation submit that 3-star hotels are not substitutable for 4-star hotels and similarly that 4-star hotels are not substitutable for 5-star hotels. As, however, several respondents also point out that some competitive pressure is exerted by hotels with a subsequent star rating, the existence of broader market segments encompassing both 4- and 5-star hotels is not excluded.

Segmentation by category

(61) A different type of hotel classification by comfort/price level is widely used among hotel operators and notably by hotel chains, as it is the classification followed by leading providers of competitive benchmarking, information services and research in the hotel industry. Chain and independent hotels are distinguished into up to six different categories (classes or scales), on the basis of their actual average room rates. In the case of chain hotels, the classification is done on a brand basis on a global chain-wide level, namely all hotels belonging to certain brand are classified in the same category, even if there is some degree of differentiation in the services they provide.

(62) As the classification into the various categories is done globally for entire brands, it is not a very accurate indicator of the comfort/price level of each specific hotel. This possible inconsistency between the category in which a specific hotel is classified and its actual quality is also reflected in the comparison between the star rating of specific hotels and the category in which these same hotels are classified. Moreover, as already indicated, even if hotel chains use this type of classification, customers are not familiar with it and rely instead on hotels' star-rating, which is widely communicated in all distribution channels.

4.2.1.2.4 Conclusion

(63) In light of the above considerations, the Commission concludes that, for the purposes of this Decision, it can be left open whether the market for hotel accommodation services should be further segmented by comfort/price level on the basis of hotels' star rating. Should narrower distinctions by star rating be considered, it can be left open whether 4-star and 5-star hotels form part of the same market. A distinction by comfort/price level on the basis of comfort classes is considered by the Commission not relevant for the purposes of this Decision.

4.2.1.3 Market for accommodation services in short stay residences

(64) In addition to the provision of hotel accommodation services, both Parties have limited serviced apartments offerings, corresponding to short stay residences. Moreover, Starwood has some "residential apartment" offerings, i.e. long-term leasehold or owned apartments in some of the hotels it operates. Marriott does not have any "residential apartment" offerings.

4.2.1.3.1 Commission's practice

(65) The Commission considered in prior decisions whether short stay residences would form part of a broader hospitality market, also including hotel accommodation services. The Commission identified some differences between the two types of services, as short stay residences are usually larger than hotel rooms and equipped with a kitchen, the services offered in addition to the accommodation are more limited, the customer base consists mainly of corporate customers and the average stay is longer than in the case of hotels. Ultimately, the Commission left the exact definition of the product market open.

4.2.1.3.2 Marriott's views

(66) Marriott submits that serviced apartments do not belong to a separate product market, as they compete not only with conventional hotel rooms, but also with other short stay properties, such as short-term lets and new entrants like Airbnb. Moreover, even if a separate market for short stay residences or serviced apartments were considered, the Parties' activities in such market are limited, as there are […] properties with a total of […] serviced apartments or residential offerings under Marriott.

4.2.1.3.3 Commission's assessment

(67) Given the limited scope of the Parties' activities in offering serviced apartments, the Transaction would not materially impact any plausible separate market for accommodation services in serviced apartments. Moreover, even if these activities were considered together with hotel accommodation services, as part of a broader market encompassing both activities, the impact of the Transaction would not materially change.

(68) The inclusion of serviced apartments in or their exclusion from the market for hotel accommodation services would therefore not affect any of the conclusions about the effects of the Transaction reached in this Decision. The Parties' serviced apartments' offerings will thus not be further analysed. However, as analysed in Section 5.2.1.3, the Parties' serviced apartments' offerings will, on a conservative basis, be added to their share on markets for hotel accommodation services.

4.2.1.3.4 Conclusion

(69) Therefore, the Commission concludes that, for the purposes of this Decision, it can be left open whether accommodation services in short stay residences and in hotels form part of the same product market.

4.2.1.4 Markets for the provision of specific services

(70) Within the framework of their overall activity as providers of hotel accommodation, the Parties also offer a number of other services. For example, some Marriott and Starwood hotels offer facilities and services for conferences, meetings and small events, fitness facilities, food and beverages through onsite restaurants and bars, gift shops, loyalty programs, etc.

(71) Such services provided on site in the various Marriott and Starwood hotels are peripheral to the Parties’ primary activity and revenue stream, which is the provision of hotel accommodation services. In that sense, meeting facilities/services, restaurant facilities, spa facilities etc. are provided by the Parties as an alternative to guests using non-hotel providers for the purchase of such services and aim at facilitating the hotel accommodation business. Moreover, asked specifically about conference and event services, the majority of the respondents having expressed an opinion in the market investigation indicate that it is ancillary to hotel operators' primary business and intended to increase the hotels' occupancy rates.

(72) In light of the above, and considering all evidence available to the Commission, the Commission concludes that the provision of additional services, e.g. for conferences and events, fitness and spa, food & beverage, loyalty programs etc. by hotel operators is ancillary to hotel accommodation services and as such will not be further analysed in this Decision.

(73) As set out in Section 5.2, the Parties' activities in relation to the provision of services beyond hotel accommodation will be taken into account in view of assessing if the Parties compete closely with each other on the markets for hotel accommodation services.

4.2.1.5 Conclusion

(74) In light of the above, and considering all evidence available to the Commission, the Commission concludes that, for the purposes of this Decision, the relevant product market is the market for hotel accommodation services, comprising hotel accommodation services provided by both chain and independent hotels.

(75) It is not necessary, for the purposes of this Decision, to conclude on a potential further segmentation of that market by comfort/price level on the basis of hotels' star rating in segments consisting of hotels with the same star rating or segments consisting of hotels with subsequent star ratings.

4.2.2 Hotel management services

4.2.2.1 Definition of hotel management services

4.2.2.1.1 Commission's practice

(76) The Commission has not, in its prior decision practice, defined a market for the provision of hotel management services.

(77) However, it has considered a separate market for the provision of real estate management services, i.e. the management and operation of real estate on behalf of its owner. Such market could be sub-segmented between the provision of management services to real estate properties for residential and for commercial use. Narrower segments based on the final destination of the properties to be managed for commercial use, such as offices, industrial properties, retail properties were also considered. Ultimately, the Commission left the exact market definition open.

4.2.2.1.2 Marriott's views

(78) Marriott submits that hotel management services are services provided to hotel owners by both hotel chain operators in the context of management agreements concluded with hotel owners as well as by other players such as management companies not affiliated with a particular hotel brand ("white label management companies"). Those services relate to the day-to-day operation of a hotel and include financial accounting, generating revenue, managing hotel-specific relations, advertising, promotion and marketing services, IT support, human resources management, executing and supervising repairs and maintenance of the hotel.

4.2.2.1.3 Commission's assessment

(79) In the Commission's market investigation, a majority of respondents agrees that the supply of hotel management services can be defined as operating a hotel for an owner typically in return for fees and/or a share of revenues (including financial accounting, generating revenue, managing hotel specific relations, advertising, promotion and marketing, IT support, managing human resources, executing and supervising repairs and maintenance).

(80) In addition, two respondents underlined the importance of brands in the definition or segmentation of hotel management services, although those two respondents reached relatively opposite conclusions.

(81) The views expressed during the market investigation on the relevance of branding or franchising services as part of hotel management services confirmed that hotel owners deciding to use external providers of hotel management services may opt for either (i) management services that include brand franchising services by the management company, as offered by hotel chains, or (ii) management services that do not include hotel branding solutions, as offered by white label management companies.

(82) Furthermore, several hotel chains active on the market for the provision of hotel management services, pointed to a possible link between, on the one hand, the use

81Case M.2825 - Fortis AG SA/Bernheim-Comofi SA; Case M.6020 - ACS/Hochtief and others.

82Form CO, paragraph 149.

83See responses to Q4 – Questionnaire to hotel owners and managers, questions 7 and 7.1 and Q1a - Questionnaire to competitors, questions 41 and 41.1.

84See response of a hotel chain to Q1a - Questionnaire to competitors, question 41.1, which disagrees with the definition of the supply of hotel management services proposed by the Commission "Because it lacks the Brand side, all the franchises services should be included" and response of a hotel owner and manager to Q4 – Questionnaire to hotel owners and managers, questions 7.4 and 7.4.1 on whether the market for the supply of hotel management services should be further segmented: "Hotel Brands should focus more on the Brands as such and leave the operation / management of the hotels to professional white label operators".

21

4.2.2.1.4 Conclusion

(83) In light of the above considerations, the Commission concludes that there is a separate market for the provision of hotel management services, defined as operating a hotel for a third-party hotel owner. Nevertheless, further assessment is needed to determine whether hotel owners regard the following hotel management services as interchangeable or substitutable: (i) services provided by hotel chains and white label management companies; and (ii) services provided for the management of hotels belonging to different star ratings or categories / classes.

(84) The Commission will thus consider whether the market for the provision of hotel management services should be segmented based on (i) the types of providers (see Section 4.2.2.2 below) and (ii) the hotel comfort/price level (see Section 4.2.2.3 below).

4.2.2.2 Distinction by type of providers (hotel chains versus white label management companies)

4.2.2.2.1 Marriott's views

(85) Marriott states that both chain operators and white label management companies offer the same types of management services to hotel owners and that from a hotel owner’s perspective, management services provided by a chain operator are substitutable with those provided by white label management companies.

(86) To support that statement, Marriott compares hotel chains and white label management companies on the basis of specific elements of the management services, that is to say the fees they charge, their international expertise, and the types of hotels they manage.

85See agreed minutes of a call with a competitor on 11 March 2016: "... has oriented its business strategy towards franchise agreements especially for brands in the midscale, upscale and upper up-scale segment, leaving management contracts for those hotels in the luxury segment where a certain expertise is required which ... does not want to delegate to a franchisee and also for hotels in emerging markets". See also response of other hotel chains to Q1a - Questionnaire to competitors, question 35 on the main barriers to entry for hotel accommodation services, question 41.4.1 on the need to further segment the market for the supply of hotel management services and question 48 on the possible variations in the types of services requested by hotel owners under management agreements: "Entry/expansion into the supply of “luxury” and “upper upscale” accommodations also requires significant management expertise and more specialised service personnel than are required for other kinds of hotel operations"; "The market should not be segmented further as there are no clear divisions. Some hotel management services providers will focus more at one end of the spectrum than the other (e.g. budget hotels vs. luxury hotels)"; and "Depending on quality of Hotel there may be some differences".

86Form CO, paragraphs 155 and 156.

87Form CO, paragraph 157.

4.2.2.2.2 Commission's assessment

(87) The responses to the market investigation questionnaires do not clearly show whether management services provided by a hotel chain, which are bundled with a license of one of the brands it owns, and management services provided by a white label management company, which are stand-alone services that the hotel owner may decide to combine (or not) with a license of a brand owned by a hotel chain, are demand substitutes.

(88) Overall, most responding hotel owners and hotel chains explained that, at the term of a management agreement, hotel management services provided by a hotel chain can be replaced by hotel management services provided by a white label management company, possibly together with a franchising agreement with a hotel brand owner.

(89) However, their views were split about the ease or difficulty associated with the replacement of a managing hotel chain by a white label management company (possibly acting together with a hotel franchisor). Indeed, the proportion of respondents indicating that such a replacement can easily take place is comparable to the proportion of respondents declaring that such a replacement is subject to specific conditions, notably: (i) an appropriate level of know-how and capabilities of the white label management company, and (ii) the ability to franchise the hotel under a brand with a comparable positioning.

(90) In addition, one hotel chain that considered that hotel management services provided by a hotel chain cannot be easily replaced by hotel management services provided by a white label management company underlined the barriers and costs for a hotel owner entailed by the switch from management by a hotel chain to management by a white label management company. The described barriers do not appear to be specific to the replacement of a hotel chain by a white label management company; they could rather also exist in a situation of change of the managing hotel chain.

(91) By contrast, the additional costs described by that hotel chain would be specific to the substitution of management services provided by a hotel chain by the combination of management services provided by a white label management company and of franchising services by a hotel brand owner.

(92) Due to the specificities of the fee structure defined under each management agreement between a hotel chain and a hotel owner, the Commission is not in a position to relevantly compare, on the one hand, the average total fees due by a hotel owner under a management agreement with a hotel chain and, on the other hand, the sum due by a hotel owner of the management fees under a management agreement with a white label management company and of the franchise fees under a franchising agreement with a hotel chain. The Commission nevertheless notes that no hotel owner mentioned the potential difference in the overall fee levels as a barrier to switching to white label management companies.

4.2.2.2.3 Conclusion

(93) In light of the above considerations, the Commission concludes that, for the purposes of this Decision, it can be left open whether the market for the provision of hotel management services should be further segmented by type of providers (hotel chains versus white label management companies).

(94) In the competitive assessment, the Commission will assess the Parties' positions on the potential segment of hotel management services provided by hotel chains, considering that (i) they do not operate on the other potential segment (services provided by white label management companies), and (ii) the Parties' combined market share would be diluted if hotel chains and white label management companies are considered as operating on the same market for the provision of hotel management services.

4.2.2.3 Distinction by hotel comfort/price level

4.2.2.3.1 Marriott's views

(95) Marriott believes that the market for the provision of management services, if any, should not be segmented on account of specific hotel categories. It supports its opinion by providing examples of hotel chain operators, such as Accor, and white label management companies, such as Westmont and Interstate, that manage hotels across all comfort/price levels (all 5 star ratings and all STR classes).

4.2.2.3.2 Commission's assessment

(96) Most respondents to the market investigation consider that the market for the supply of hotel management services should not be further segmented.

(97) Two hotel chains referred specifically to management services provided in the luxury and upper-upscale segments or in high quality scales. However, those references seem to aim at (i) qualifying the level of substitutability of management services provided by hotel chains, and (ii) insisting on the in-depth expertise required to operate high-end hotels, without implying however that such operations would constitute a separate market from operating lower-end hotels.

(98) No hotel owner mentioned that the hotel management services they demand or the level of hotel management fees they owe to their providers would depend on the comfort/price level of the managed hotels.

4.2.2.3.3 Conclusion

(99) In light of the above considerations, the Commission concludes that, for the purposes of this Decision, the market for the provision of hotel management services should not be further segmented by hotel comfort/price level.

(100) However, in assessing the competitive impact of the Transaction, the Commission will take into account the fact that Marriott and Starwood operate mainly high-end hotels (i.e. mainly 4- and 5-star hotels according to star ratings or luxury, upper upscale and upscale hotels according to STR classes).

94Form CO, paragraph 160.

95See responses to Q4 – Questionnaire to hotel owners and managers, questions 7.4 and 7.4.1 and Q1a - Questionnaire to competitors, questions 41.4 and 41.4.1.

96See response of a number of hotel chains to Q1a - Questionnaire to competitors, question 41.3.1, on whether the hotel management services provided by a hotel chain can easily be replaced by hotel management services provided by a white label management company, possibly together with a franchising agreement with a hotel brand owner: For instance, "[...] considers that the provision of hotel management services by a "white label" operator may provide a viable means of operating an "up-scale" or lower-end property and, in fact, such operations are rather common. However, reliance on such "white label" services in the "luxury" and "upper-upscale" segments is much more risky, given the more specialised capabilities required to operate such a multi-faceted facility and the more substantial investments at risk. Accordingly, [...] considers that such management options are less viable/common for "luxury" and "upper-upscale" properties"; "(...) However, the higher up the quality scale one operates, the more such soft factors as corporate culture and differentiated/ personalised service delivery play an integral part in brand differentiation and achieving brand premium. this is extremely difficult for white label operator to replace/ copy, especially if their business model is build around offering management services under different brands".

97This does not exclude the possibility of variations in the level of fees charged by hotel chains based on the brands under which the hotels will be operated.

25

4.2.2.4 Conclusion

(101) In view of the above, and considering all evidence available to the Commission, the Commission considers that the relevant product market is the market for the provision of hotel management services and that the question of its segmentation by type of providers (hotel chains versus white label management companies) can be left open, since no competition concerns arise on any plausible product market definition.

4.2.3 Hotel franchising services

4.2.3.1 Commission's practice

(102) The Commission has in its prior decision considered a market for hotel franchising services, leaving the exact market definition open.

4.2.3.2 Marriott's views

(103) Marriott submits that franchise agreements essentially consist of licenses of industrial or intellectual property rights relating to trademarks or systems and know-how. In the hotel industry, the hotel is operated under the franchisor's brand name and the franchisor provides services to the third-party owner / manager against payment of fees by the third-party owner to the franchisor.

4.2.3.3 Commission's assessment

(104) In the Commission's market investigation, the Commission proposed to define hotel franchising as the issuing by a company (the "franchisor") of a contract authorising an unrelated company (the "franchisee") to use a specific name and logo, purchased for an annual fee plus "royalties" usually based on a percentage of sales. Franchisees share such benefits as brand-name identity, corporate image advertising, centralised reservation systems, corporate training programs and volume purchasing. A majority of respondents agrees with that definition.

(105) In addition, most respondents having expressed an opinion consider that the market for the provision of hotel franchising services should not be further segmented.

(106) Furthermore, no hotel owner mentioned that the hotel franchising services they demand or the level of hotel franchise fees they owe to their providers would depend on the typology or on the comfort/price level (including on the star rating or on the category / class) of the franchised hotels.

(107) Therefore, the Commission considers that, for the purposes of this Decision, the market for the provision of hotel franchising services should not be further segmented by hotel comfort/price level.

4.2.3.4 Conclusion

(108) In view of the above, and considering all evidence available to the Commission, the Commission concludes that the relevant product market is the overall market for the provision of hotel franchising services.

(109) However, in assessing the competitive impact of the Transaction, the Commission will take into account the fact that Marriott and Starwood franchise mainly high-end hotel brands (i.e. mainly 4- and 5-star hotel brands according to star ratings or luxury, upper upscale and upscale hotel brands according to STR classes).

4.3 Relevant geographic markets

4.3.1 Hotel accommodation services

4.3.1.1 Commission's practice

(110) In prior decisions, the Commission has left open the exact geographic scope of the market for hotel accommodation services. It has however noted that the relevant geographic market presented both national and local characteristics. The market was considered national because the structure of supply may vary from one market to another since the hotel industry is linked to national economic trends, whereas the conditions for competition are homogeneous at national level. In addition, the market was considered local, because a second degree of competition exists at a local level, the primary criterion for the choice of a hotel being its location. Cities can be considered as local markets for hotels as one of the main features of the hotel sector is its individual city character: customers select hotels in the city where they stay.

4.3.1.2 Marriott's views

(111) According to Marriott, the Commission should assess the Transaction by reference to cities. Indeed, from a demand-side perspective, for the vast majority of customers, hotels located in different cities in the same country are not substitutable with each other. From a supply-side perspective, in setting their prices, hotels will consider their product and price offerings relative to their competitors within the city and with respect to city-wide demand conditions. Marriott adds that administrative boundaries represent the most accurate approach to define a city, as they reflect the scope of each individual city irrespective of its size and layout in the most objective way. Further, administrative boundaries appear to correspond to the definition of a city found on OTAs, thereby also reflecting the way in which customers look for hotel accommodation services. Marriott is also of the view that there is no national (or wider) competition as all customer groups tend to make their choices on a local basis.

4.3.1.3 Commission's assessment

(112) The majority of respondents to the market investigation submit that hotels primarily compete with each other at a local level for hotel accommodation services.

(113) From the supply side, the way in which hotel operators benchmark themselves and analyse the market, for instance in view of setting prices, gathering market data etc., is also indicative of how they view the geographic scope of the market. Indeed, the competitive sets designed by Marriott and Starwood hotels include a number of competing hotels in the same location and not in different locations within the same country. Similarly, when asked about the way prices are set at the various hotels, the majority of respondents indicate that, even if some pricing guidelines are set centrally in the case of hotel chains, the final price is set at hotel level. Moreover, when setting prices, the majority of respondents explain that they take into account the market prices for hotels of the same comfort/service level in the same location and that the prices of hotels in the vicinity are a very important criterion in setting the rates for the supply of hotel accommodation services by a specific hotel.

(114) From the demand side as well, the majority of respondents having expressed an opinion in the market investigation indicate that even if corporate contracts have a broad, often worldwide scope and are negotiated at chain level, they are for the purpose of the tendering procedure organised on a city-by-city basis, whereby specific terms, such as the participating hotels, the pricing, etc., are dealt with at local level.

105 Form CO, paragraphs 197 et seq.

106 See responses to Q1a – Questionnaire to competitors, question 14, Q1b – Questionnaire to independent and franchised competitors, question 13, Q2 – Questionnaire to corporate customers, question 10, Q3 – Questionnaire to consumer associations, question 10, Q5 – Questionnaire to travel agencies and tour operators, question 12.

107 Competitor sets of the Parties' hotels, as extracted from their revenue optimisation tools and annexed to their management or franchise contracts; Form CO, paragraph 865 and following and Annexes 8.1.a.1, 8.1.a.2 and 8.1.b.1.

108 See responses to Q1a – Questionnaire to competitors, question 15, Q1b – Questionnaire to independent and franchised competitors, question 14, Q5 – Questionnaire to travel agencies and tour operators, question 13.

109 See responses to Q1a – Questionnaire to competitors, question 16, Q1b – Questionnaire to independent and franchised competitors, question 15.

110 See responses to Q1a – Questionnaire to competitors, question 18, Q1b – Questionnaire to independent and franchised competitors, question 7.

28

111 el. Indeed, customers explain that in view of concluding corporate contracts, they first identify the locations in which they are interested and subsequently invite, directly or through the hotel chain headquarters, hotels in these locations to submit an offer for their accommodation services. Moreover, a large number of respondents point out that location is among the key drivers of customers' selection.

(115) As to the precise delineation of the local market, most respondents having expressed an opinion in the market investigation indicate that this could be either a city or a city district, in particular in the case of larger cities like London, Paris, etc. Moreover, industry reports, such as the STR Global's STAR Report, analyse the market primarily on the basis of cities based on their administrative borders, as well as according to smaller districts within a city ("tracts"), notably as far as larger cities are concerned.

(116) Such narrower segmentation, however, does not appear to adequately reflect the competitive relationship between hotels located in the same city. Indeed, by sub-segmenting on that basis, hotels falling under different tracts or belonging to different districts may be closer to each other than hotels within the same tract or district. In the case of London for example, if a narrower market for North Central London were considered, hotels in Camden Town and hotels in Marylebone would be considered as part of the same geographic market, even though within a distance of approximately 3 km, whereas hotels in Marylebone and Mayfair at 1 km distance would belong to separate markets. Hotels located in neighbouring city districts appear thus to exert competitive pressure on each other. Therefore, even if such narrower segments were to be considered, a chain of substitution would exist between hotels located in neighbouring districts and the total market would again amount to a

See responses to Q1a – Questionnaire to competitors, questions 26 and 27, Q1b – Questionnaire to independent and franchised competitors, question 25, Q2 – Questionnaire to corporate customers, questions 11 and 12.

See agreed minutes of a call with a customer of 2 March 2016, a call with a customer of 4 March 2016, a call with a customer of 3 March 2016.

See responses to Q1a – Questionnaire to competitors, question 5, Q1b – Questionnaire to independent and franchised competitors, question 4, Q2 – Questionnaire to corporate customers, question 2.2, Q3 – Questionnaire to consumer associations, question 2, Q5 – Questionnaire to travel agencies and tour operators, question 4. Even though the question did not offer location as an option, several respondents added in their comments that it constitutes one of the main drivers for consumers' selection of hotel accommodation provider.

See responses to Q1a – Questionnaire to competitors, question 14.1, Q1b – Questionnaire to independent and franchised competitors, question 13.1, Q2 – Questionnaire to corporate customers, question 10.1, Q3 – Questionnaire to consumer associations, question 10.1, Q5 – Questionnaire to travel agencies and tour operators, question 12.1.

Form CO, Annex 6.2.1.1. Besides, following the STR segmentation of the market in tracts, the Parties' market shares would be different in some tracts than on a city-basis. These data, which were based on comfort classes and not on star rating, were explained and reconciled by the Parties with their own data used in the Form CO.

Form CO, Annex 6.3.1.3.

Form CO, Annex 6.2.1.1.

broader city-wide market for hotel accommodation services.As set out in Section 5.2.3.1.2, however, the potential higher competitive pressure exerted between hotels located in close proximity will be taken into account when assessing whether the Parties compete closely with each other on the markets for hotel accommodation services.

(117) The case of city airports may however be somewhat different, as hotels in the airport area are located in close proximity to the airports, which in many instances are in the outskirts of their respective city. Moreover, the consideration of the relevant geographic market on the basis of an area's administrative limits would likely not reflect the competitive conditions also in the case of non-city destinations. In the cases of resorts, islands, etc., the administrative limits of a city may not reflect the scope of the market for hotel accommodation services. Elounda, for example, in which the Parties' activities overlap, is a small fishing town in the Lasithi region in Crete. Numerous other similar towns exist in the same region and in Crete as a whole and hotels situated in those towns are in competition with those in Elounda for guests visiting Crete.

(118) In view of the above, and considering all evidence available to the Commission, the geographic market for hotel accommodation services appear to be local.

(119) The exact delineation of the local market may however be left open, including the possibility of city-wide markets or markets covering a broader e.g. resort area, since no competition concerns arise on any plausible geographic market definition. Similarly, whether cities and their respective airport areas shall be considered as part of the same geographic market may also be left open, as no competition concerns would arise, irrespective of the exact delineation of the geographic market.

118Commission Notice on the definition of relevant market for the purposes of Community competition law, paragraph 57.

119See agreed minutes of a conference call with a competitor of 10 March 2016.

Form CO, paragraph 511; even though there are [number of hotels] hotels operating under a Marriott and a Starwood brand respectively in Elounda, [all] hotels are franchised. As analysed in Section 5.2.1.2 below, franchised hotels are operated independently and not considered as part of the Parties' activity on a market for hotel accommodation services. Therefore, the franchising of [number of hotels] hotels in Elounda does not give rise to an overlap between the Parties' activities in the local market.

Even if national markets were to be considered, the Transaction would not give rise to serious doubts as to its compatibility with the internal market. More concretely, the Transaction would result in combined market shares of more than 30% only on the Polish market for 5-star hotels. However, the Parties are not particularly close competitors on the Polish market and the majority of respondents to the market investigation indicate that sufficient competition would remain in these markets also post-Transaction. See responses to Q1a – Questionnaire to competitors, question 67, Q1b – Questionnaire to independent and franchised competitors, question 51, Q2 – Questionnaire to corporate customers, question 23, Q3 – Questionnaire to consumer associations, question 18, Q5 – Questionnaire to travel agencies and tour operators, question 30.

The Commission concludes that, for the purposes of this Decision, the market for hotel accommodation services should not be further segmented by city districts.

4.3.2 Hotel management services

4.3.2.1 Commission's practice

(121) The Commission has not in its prior decision practice defined a market for the provision of hotel management services.

(122) Concerning the potential market for the provision of real estate management services and all its sub-segments, the Commission has in its prior decision practice left the geographic market definition open, notably whether it is national or local.

4.3.2.2 Marriott's views

(123) If there were a separate market for the provision of hotel management services, Marriott considers that the narrowest possible scope would be EEA-wide because management services are being provided across the EEA (and often globally) by the relevant hotel chains, as well as by white label management companies. A potential geographic market which would be narrower than EEA would not reflect market reality as the majority of both chain operators and white label management companies are seeking management opportunities and provide their services at least across the EEA and most of them globally.

(124) Indeed, Marriott considers that the most fundamental aspects of the competition conditions on a potential market for the provision of hotel management services are broadly similar at a worldwide level. Over the past two decades, Marriott has observed an increasingly competitive worldwide marketplace for the provision of management services to hotel owners. Equally, there is a global demand for the provision of management services.

(125) In addition, the types of services offered under a management agreement are broadly similar across the globe. Likewise, the fee structure of Marriott's management agreements is broadly uniform worldwide, with the exception of the USA.

4.3.2.3 Commission's assessment

(126) To obtain the general views of market participants on the plausible geographic markets for the provision of hotel management services, the Commission asked the Parties' competitors as well as hotel owners and white label management companies to identify the geographic area on which hotel management companies primarily compete with each other for the supply of hotel management services.

122See notably Case M.2825 - Fortis AG SA/Bernheim-Comofi SA; Case M.6020 - ACS/Hochtief.

Form CO, paragraph 210.

See Marriott's answer of 07 June 2016 to RFI 12.

That general question yielded mixed results. The most frequent answer was "worldwide", but the results of the market investigation do not enable to totally exclude "locally", "nationally" or "it depends". In addition, most respondents having expressed an opinion considered that hotel management companies also compete with each other in other geographic areas for the supply of hotel management services.

With regard to the plausibility of a local market, the Commission is of the view that the minority of hotel owners taking the view that hotel management companies primarily compete locally for the supply of hotel management services explain their choice by reference to a different market, i.e. the market for hotel accommodation services. They notably refer to the local competition between hotels, which may encourage multi-property hotel owners to diversify their portfolio of management companies.

The influence of the functioning of the market for hotel accommodation services is also to be found in the explanations of hotel owners that chose "it depends" as a reply to the question on the geographic area on which hotel management companies primarily compete. One hotel owner notably explained that the geographic scope of the competition for the provision of hotel management services depended on the size of the local market (city).

However, no hotel owner established any correlation between the scope of the competition on the market for hotel accommodation services and the scope of the competition on the market for the provision of hotel management services. In particular, no hotel owner mentioned that the market for the provision of hotel management services was characterised by differences in price or demand at local level (there was, for instance, no preference for a local management company or a local brand).

(135) The Commission notes that the Parties' market shares vary significantly from one EEA country to another and that the competitive landscape shows certain differences, notably with the presence of national players in some countries. The differences in the Parties' and their competitors' market shares are confirmed by the market reconstruction undertaken by the Commission.

(136) Nevertheless, the Commission considers that those national variations result mainly from the "weight of the past" rather than from heterogeneous conditions of competition. More particularly, the structure of the market for the provision of hotel

agement company"; "The level of management fees tend to be standard, even if they may vary depending on the typology and characteristics of the hotel, specially in very special assets."

130See Commission Notice on the definition of the relevant market, paragraph 8.

See responses to Q1a – Questionnaire to competitors, questions 47 and 47.1. For example: "Finally, the conditions of competition are homogeneous as hotel management companies offer quite similar services to the hotel owners. The competition does not rely on the offered service but on the experience and the reputation of the hotel management companies."

See responses to Q4 – Questionnaire to hotel owners and managers, questions 13, 13.1.

See Form CO, paragraphs 810-817 and Annex 6.3.2. As an example, taking account of management services provided by hotel chains only, Starwood has a [5-10]% market share in Belgium, where the Parties have not identified a strong competitor, and a [0-5]% market share in France, where Groupe du Louvre and Accor notably have a strong presence.

134See Section 5.3.1 of this Decision.

135See Commission Notice on the definition of the relevant market, paragraph 29.

management services seems to be inherited, at its early stage of development, from the structure of the market for hotel accommodation services, for the following two reasons.

(137) First, hotel chains need to have demonstrated their capacity to successfully self-manage hotels before being able to compete successfully for the provision of hotel management services to third-party owners. Therefore, strong players on the market for hotel accommodation services in a certain country are likely to capitalise on their strength and grow into strong players on the market for the provision of hotel management services at national level, thus enjoying higher market shares in their domestic market before expanding to other markets.

(138) Second, the market for the provision of hotel management services results, at least partly, from the development of an asset-light business model in the hotel industry and the trend towards the divestiture by hotel chains of the assets they own. Since a hotel chain that engages in the divestiture of the hotels it owns is likely to continue, at least initially, managing its divested hotels, the heterogeneous presence of a hotel chain throughout the EEA, and a strong domestic market share, may simply derive from the heterogeneous geographic distribution of its hotels.

4.3.2.3.2 Pricing and basic demand characteristics

(139) First, Marriott's pricing and contracting practices are broadly similar across the EEA. Starwood's pricing and contracting practices are consistent not only across the EEA States, but also across the broader Europe, Africa and Middle East region and worldwide.

(140) Importantly, most respondents to the market investigation questionnaires having expressed a view consider that the types of services requested by hotel owners under management agreements and the level of management fees do not vary significantly from one EEA country to another.

(141) In addition, respondents mentioning hotel owners' possible national preferences acknowledge that those preferences would have an effect on (i) the level of expertise required from hotel management services rather than on their place of establishment, and (ii) the focus of the management services to be provided rather their nature. Therefore, any national preferences would not lead to the differentiation of the hotel management services and of the key provisions of the management agreements between EEA countries.

This is illustrated by the explanation given by a hotel chain to support its response to the questions on the geographic area on which hotel management companies compete with each other for the supply of hotel management services (see Q1a – Questionnaire to competitors, questions 46.1 and 46.2.1): "Management companies compete globally but they tend to expand gradually, reaching a critical mass of rooms operated in the same region, meaning that a management company tends to concentrate the operation of hotels in a certain country and the near countries, in order to share/cluster services and maximize the synergies of the operation. Following that strategy the management company of course could expand to various different regions and countries worldwide."

Form CO, paragraphs 217-225.

Form CO, paragraphs 226-230.

See responses to Q4 – Questionnaire to hotel owners and managers, questions 14, 14.1, 15 and 15.1 and Q1a – Questionnaire to competitors, questions 48, 48.1, 49 and 49.1. For example: "The level of management fees do not depend on the EEA country considered. However, it may vary depending on the negotiation between the hotel owner and the hotel management company."

4.3.2.4 Conclusion

(150) In view of the above, and considering all evidence available to the Commission, the Commission considers that, for the purpose of this Decision, the market for the provision of hotel management services is at least EEA-wide.

(151) The Commission will assess the effects of the Transaction at both EEA- and worldwide levels.

4.3.3 Hotel franchising services

4.3.3.1 Commission's practice

(152) The Commission has in its prior decision considered that the market for hotel franchising services presents supra-national characteristics, ultimately however leaving the exact market definition open.

4.3.3.2 Marriott's views

(153) Marriott submits that any market for the provision of hotel franchising services would be worldwide and provides examples of brands from Marriott, Starwood and competitors that are franchised on a global basis.

(154) Marriott acknowledges that there are differences in shares by hotel brands from country to country. Nevertheless, it argues that those differences are due to the time needed to expand a hotel brand across the world in an intensely competitive marketplace.

(155) In addition, Marriott considers that the most fundamental aspects of the competition conditions on a potential market for the provision of hotel management services are broadly similar at a worldwide level. There is a global demand for the provision of hotel franchising services. The types of services offered under a franchise agreement and its main terms and conditions, including franchise fees, are broadly similar across the globe.

4.3.3.3 Commission's assessment

(156) To obtain the general views of market participants on the plausible geographic markets for the provision of hotel franchising services, the Commission asked the Parties' competitors as well as hotel owners and white label management companies to identify the geographic area on which hotel franchisors primarily compete with each other for hotel franchising, in terms of (i) brand awareness (i.e. area where the competitors' brands are known), and (ii) presence (area where competitors have a hotel already operating under their brands).

147Case M.2510 – Cendant/Galileo, paragraph 17.

148Form CO, paragraph 232.

149Form CO, paragraph 235.

150See Marriott's answer of 07 June 2016 to RFI 12.

(157) The majority of respondents indicate that hotel franchisors compete with other franchisors whose brands are known world-wide, followed by those considering that "it depends". The opinions of respondents were divided as to whether hotel franchisors also compete with each other, in terms of brand awareness, in other geographic areas.

(158) There was no consensus on whether hotel franchisors compete with each other based on their presence on certain geographic area, as the majority of respondents answered that "it depends", followed by those – predominantly hotel chains – considering that they compete with other franchisors having a world-wide network of hotels operating under their brands, and those considering that they compete with franchisors having a hotel operating under their brands in the specific location in which the hotel to be franchised is located. Finally, the responses are also divided as to whether hotel franchisors also compete with each other for hotel franchising, in other geographic areas in terms of presence.

(159) First, with regard to the plausibility of a local market, the Commission is of the view that the respondents having taken the view that hotel franchisors primarily compete locally for hotel franchising explain their choice by reference to a different market, i.e. the market for hotel accommodation services.

(160) However, no hotel owner established any correlation between the scope of the competition on the market for hotel accommodation services and the scope of the competition on the market for the provision of hotel franchising services. In particular, no hotel owner mentioned that the market for the provision of hotel franchising services was characterised by differences in price or demand at local level (there was, for instance, no preference for a local brand).

151See responses to Q1a - Questionnaire to competitors, question 42.1 and to Q4 – Questionnaire to hotel owners and managers, question 8.1.

See responses to Q1a - Questionnaire to competitors, questions 42.2 and Q4 – Questionnaire to hotel owners and managers, question 8.2. For example: "It depends because the geographic market could vary. In fact, the geographic market will be national for those franchisor having hotels already operating under their brands in the country of the hotel to be franchised, or it will be EEA-wide for those franchisors having a EEA-wide network of hotels already operating under their brands, etc." and "It would depend on the size, star rating, existing network and global aspiration of the franchisors, though most franchisors tend to at least have aspirations of national reach( versus only local)."

See responses to Q4 – Questionnaire to hotel owners and managers, question 8.2.1. For example: "Hotels generally compete with each other in the area around where they are located since customers select hotels because they want to visit the area nearby" and "A hotel's total revenue will comprise customers from a range of demographics." See also responses to Q1a - Questionnaire to competitors, question 42.2.2.1. For example: "Each franchise normally applies to a single hotel, which trades in a local trading area, and competition for that franchise typically includes only hotel operators who do not already have a comparably rated hotel in that trading area. While franchisors may compete with each other more broadly, e.g. to establish a viable/competitive footprint across the major destinations in Europe, competition for individual franchises nonetheless is ultimately local, with potential franchisors being likely to pursue new opportunities in light of their existing properties/commitments."

If any, price differences would be linked to each franchise agreement or to the relationship between the franchisor and the franchisee (hotel owner). See for example responses to Q4 – Questionnaire to hotel owners and managers, question 11.1: "However, any differential in fees is more likely to now be driven by strategic commercial objectives, as agreed by the franchisor and franchisee" and "They are usually and normally standard Fees, which only vary when there is Multiple Franchise Agreement in place."

(161) In light of the above considerations, the Commission concludes that, for the purposes of this Decision, the market for the provision of hotel franchising services is wider than local.

(162) Second, in order to determine whether the market for the provision of hotel franchising services is national or wider than national, in particular at least EEA-wide, the Commission has assessed the homogeneity of the conditions of competition across the EEA.

(163) The general question in the market investigation questionnaires on whether the conditions of competition on the hotel franchising market are homogeneous in the EEA triggered divided opinions. Most hotel chains having expressed a view consider that the conditions of competition are homogeneous. On the contrary, most hotel owners and white label managers having expressed a view consider that the conditions of competition are significantly different between EEA countries.

(164) The Commission will therefore rely for its assessment on different factors that, in line with the Commission Notice on the definition of the relevant market, should be taken into account for the definition of the geographic scope of the market for the provision of hotel management services.

4.3.3.3.1 Distribution of market shares between the Parties and their competitors

(165) As acknowledged by Marriott, the market shares of the Parties vary significantly from one EEA country to another. The competitive landscape shows certain differences, notably with the presence of national players in some countries. The differences in the Parties' and their competitors' market shares are confirmed by the market reconstruction undertaken by the Commission.

(166) Nevertheless, the Commission considers that those national variations result mainly from the "weight of the past" rather than from heterogeneous conditions of competition. More particularly, the structure of the market for the provision of hotel franchising seems to be inherited, at its early stage of development, from the structure of the market for hotel accommodation services.

155See Commission Notice on the definition of the relevant market, paragraph 8.

See responses to Q1a – Questionnaire to competitors, questions 43 and 43.1. For example: "Finally, the conditions of competition are homogeneous as franchisors offer quite similar services to franchisees. The competition does not rely on the offered service but on the brand awareness."

See responses to Q4 – Questionnaire to hotel owners and managers, questions 9 and 9.1. For example: "the market conditions in some countries require more heated and competitive competition with varied Terms & Conditions".

See Form CO, paragraphs 818-827 and Annex 6.3.4. As an example, Starwood has a [0-5]% market share in France, where the Parties have identified a strong competitor established in France (Groupe du Louvre), and a [20-30]% market share in Greece, where the market is dominated by foreign players.

159See below, Section 5.3.1 of this Decision.

160See Commission Notice on the definition of the relevant market, paragraph 29.

(174) As to the degree of maturity of the market for the provision of hotel franchising services in different countries, it does not delineate different markets, since the relative size of the market for the provision of hotel franchising services in a certain country has no impact on the conditions of competition for the award of the corresponding franchise agreements. As an example, according to the market reconstruction, the same hotel chain (the French group Accor) is the market leader in the provision of

166See responses to Q4 – Questionnaire to hotel owners and managers, question 11.1 and to Q1a – Questionnaire to competitors, question 45.1. For example: "There are two types of franchise fees: (i) the trademark fees and (ii) the billable service fees. Trademark fees may vary in function of the brand awareness. For instance, […]’s brands are well-known in [name of the relevant countries]. In these [number of] countries, trademark fees paid by franchisees may be higher than in other countries. Regarding billable services (i.e. access to the distribution system, marketing task, loyalty program and IT support), fees are similar in all the EEA countries."

167See responses to Q1a – Questionnaire to competitors, question 43.1. For example: "The conditions of competition are significantly different between EEA countries because the different local legal policies as well as the cultural aspects"; and "There is no single European or EEA franchising law. Some EU countries have formal franchise disclosure laws and laws which require certain terms to be in the agreements. For example, France, Italy, Spain, Sweden, Romania, and Belgium each have distinct disclosure laws. Otherwise each franchisor will take into account a range of considerations including registration and licensing of intellectual property, the local judicial system and enforceability of agreements. Franchisors will also take into account tax regimes and applicable withholding taxes."

168See responses to Q1a – Questionnaire to competitors, question 43.1. For example: "The conditions of competition in the hotel franchising market vary significantly across the EEA, with factors such as franchise model maturity and the popularity or acceptance of chain brands versus independent hotels varying between areas."; and "From a commercial perspective, there are different maturity levels across EEA countries in terms of recognition of franchising as a business model. Franchising is more common for example in Germany, France, and the UK than it is in Latvia and Lithuania."

(175) Besides, some brands (e.g. MOXY) have expanded their presence across the EEA or worldwide.

(176) In view of the above, and considering all evidence available to the Commission with regard to (i) the pricing of hotel franchising services, (ii) the characteristics of the demand in franchising services, and (iii) the profile of companies competing for the branding of the most attractive properties throughout the EEA, the Commission considers that the national discrepancies in the conditions of competition between EEA countries are not significant enough to question their overall homogeneity across the EEA.

(177) Moreover, most large hotel chains having responded to the market investigation questionnaires consider that they compete with each other for the provision of hotel franchising services mainly at worldwide level.

(178) There are therefore indications that the market for the provision of hotel franchising services is likely to be wider than EEA-wide and even worldwide. However, for the purpose of this Decision, the question of whether it is EEA-wide or worldwide can be left open because the assessment of the Transaction would not lead to any competition concerns under either market definition.

4.3.3.4 Conclusion

(179) In view of the above, and considering all evidence available to the Commission, the Commission considers that, for the purpose of this Decision, the market for the provision of hotel franchising services is at least EEA-wide.

(180) The Commission will assess the effects on the Transaction at both EEA- and worldwide levels.

5 COMPETITIVE ASSESSMENT

5.1 Introduction

(181) Under Article 2(2) and (3) of the Merger Regulation, the Commission must assess whether a proposed concentration would significantly impede effective competition

See Form CO, paragraph 234.

170See responses to Q1a – Questionnaire to competitors, questions 42.1, 42.1.1, 42.2 and 42.2.1. For example: "[…] considers that franchises commanding pan-European or worldwide brand recognition can offer substantially greater benefits to potential franchisees than franchisors of national brands can do. A potential franchisee normally will try to obtain the strongest branding (and related operational support) that it can. Accordingly, the more typical franchisee would accept a franchise from a franchisor whose brands have limited (national) recognition only if it were unable to acquire more prominent and well-supported branding"; and "The franchisors with world-wide network compete between each other with already operating hotel under their brands".

(182) According to the Commission's Horizontal Merger Guidelines concentrations may significantly impede effective competition, in particular by creating or strengthening a dominant position by eliminating important competitive constraints on one or more firms, which consequently would have increased market power, without resorting to coordinated behaviour (non-coordinated effects).

(183) The assessment of the compatibility of the Transaction with the internal market will focus on non-coordinated horizontal effects in the markets for hotel accommodation where the Parties' activities overlap (Section 5.2), as well as on the markets for hotel management services (Section 5.3), and hotel franchising services (Section 5.4).

(184) The Horizontal Merger Guidelines set out a number of factors that the Commission applies when assessing whether a notified concentration will give rise to horizontal non-coordinated effects. To reach the conclusion that a notified concentration is likely to lead to such effects, not all of those factors need to be present. Moreover, those factors are not exhaustive.

5.2 The markets for hotel accommodation services

5.2.1 Methodology for the estimation of market shares on markets for hotel accommodation services

(185) Generally, a merger giving rise to non-coordinated effects would significantly impede effective competition by creating or strengthening the dominant position of a single firm, one which, typically, would have an appreciably larger market share than the next competitor post-merger. According to well-established case law, very large market shares – 50 % or more – may in themselves be evidence of the existence of a dominant market position.

(186) As indicated in Section 4.1, the Parties are active on the market for hotel accommodation services through owned or leased, managed and franchised hotels.

5.2.1.1 Inclusion of managed hotels in the calculation of market shares

(187) In its prior practice, the Commission has considered that the provision of hotel management services to a hotel owner on a contractual basis may lead to a situation of joint control over the hotel by the hotel owner and the hotel manager.

(188) Marriott submits that the Parties usually enter into standardised management contracts, which provide that they are engaged to supervise, direct and control the operation of the respective hotel with the aim to achieve long-term profitability while maintaining brand standards. Marriott or Starwood in their function as hotel managers are granted a broad authority by the hotel owner to control all major aspects of the operation of the hotel, including setting rates, collection of revenues, preparation of books and financial reporting, and performance of repairs and maintenance. Managerial discretion is subject to the review or approval by the owner in certain major areas, such as the annual operating and capital expenditure budgets.

(189) According to the management contracts, the performance and compensation of Marriott or Starwood as hotel managers are determined on the basis of the performance of the managed hotel and not on the basis of Marriott’s or Starwood's overall performance. The Parties' management fees consist of a percentage of the hotel's gross annual revenues and a percentage of the hotel's annual operating profit. The management contracts the Parties enter into have a typical duration of […] years.

(190) The majority of competitors and hotel owners and managers responding to the market investigation also describe hotel management contracts as hotel-specific, providing for the use of centralised services and systems for the operation of the managed hotel, the manager's basic and incentive fees, the rights and obligations of each side, e.g. in relation to maintenance, performance, the control exerted by the owner, etc., a long duration of typically 10 to 20 years and the conditions for termination.

(191) Hotel managers thus appear to have a determining role on the operation of managed hotels and the development and implementation of its business strategy. As a result, Marriott and Starwood in their function as hotel managers also exert significant influence over the hotels they manage.

(192) Therefore, the Commission concludes that managed hotels are most likely jointly controlled by Marriott or Starwood. Therefore, managed hotels have been included in the calculation of the Parties' market shares in EEA markets, as more accurately reflecting their relative market position.

176See M.5164 - MSREF/IHG/Crowne Plaza Wiesbaden.

177There may be some limited variation in terms in the case of older Starwood management contracts, see Annex 2.2.b.1.

178Form CO, paragraph 17, Annexes 2.2.a.1 and 2.2.b.1.

179Form CO, Annexes 2.2.a.1, 2.2.a.2, 2.2.b.1 and 2.2.b.2.

180See responses to Q4 – Questionnaire to hotel owners and managers, question 24.

5.2.1.2 Non-inclusion of franchised hotels in the calculation of market shares

(193) According to the Consolidated Jurisdictional Notice, "franchising agreements as such do not normally confer control over the franchisee's business on the franchisor". The Commission has looked into hotel franchising in prior cases, ultimately however leaving open whether franchised hotels should be added to the calculation of the franchisors' market shares.

(194) Marriott submits that, with some negotiated variations and adjustments for brand-specific differences, the Parties enter into standardised agreements with their franchisees. Through these agreements, franchisees acquire access to, respectively, the Marriott or Starwood brand names, trademarks and operating systems in return for a fee and on condition that they respect certain minimum brand standards. The franchisee is responsible for the day-to-day operation of the hotel, for significant strategic decisions, including price setting, budget and business plan, for the staffing, maintenance and investments of the hotel.

(195) Contrary to managed hotels, Marriott and Starwood in their function as hotel franchisors do not have decision-making authority in relation to the operation of the hotel. Against the background of ensuring that brand standards and the brand reputation are respected, franchisors have specific rights, such as the right to approve the hotel manager, material hotel renovations or change of ownership of the hotel. For their franchising services, the Parties are typically remunerated through a franchise application fee, a royalty fee, depending on the hotel's gross sales, and ancillary fees that cover miscellaneous expenses, such as the cost of the on-line reservations platform and above property sales and marketing costs. The average duration of the Parties' franchise contracts is […] years.

(196) The majority of competitors and hotel owners and managers responding to the market investigation also indicate that the key terms of hotel franchise contracts consist in the brand designation and brand standards, the determination of franchisor's fees, the rights and obligations of each side, e.g. franchisor's audits to ensure brand standards, services to be provided by the franchisor, performance measures, a duration of typically at least 10 years and the conditions for termination.

(197) The majority of managers of franchised hotels responding to the market investigation indicate that the pricing strategy of the hotel is decided at the level of the hotel by them as hotel managers. Similarly, hotel chain operators, corporate customers and

181See among others, Case M. 2197 Hilton / Accor / Forte / Travel Services JV, Case M.7705 – Benson Elliot / Walton Street/ Starwood / Hotel Portfolio.

182Form CO, Annexes 2.2.a.1 and 2.2.b.1.

183[Reference to the exercise of the Parties' right of approval in practice]; Form CO, Annexes 2.2.a.1 and 2.2.b.1.

184Form CO, Annexes 2.2.a.1 and 2.2.b.1.

185See responses to Q1a – Questionnaire to competitors, question 51, Q4 – Questionnaire to hotel owners and managers, question 17.

186See responses to Q1b – Questionnaire to independent and franchised competitors, question 14.

(198) The Commission concludes on the basis of the market investigation results that the franchised hotels are operated independently by their hotel managers and not centrally by the hotel chain itself. The Parties' franchised hotels are therefore competing with other Marriott or Starwood hotels having a different manager on the same market. This is also confirmed when looking into franchised hotels competitive sets, as Marriott and Starwood hotels are included in the competitive sets of Marriott and Starwood franchised hotels respectively.

(199) In light of the above considerations, Marriott and Starwood do not have a determining role on the operation of their franchised hotels and the development and implementation of the latter's business strategy. As a result, the Parties, in their function as hotel franchisors, do not control their franchised hotels, which are instead operated by their respective hotel owner and/or manager and are thus in competition with other hotels of the same chain. Therefore, Marriott and Starwood's franchised hotels have not been included in the calculation of the Parties' market shares in EEA markets, as this would not accurately reflect their relative market position.

5.2.1.3 Calculation of market shares

(200) Marriott submits that there are no reliable objective reports allowing it to provide market share data on the basis of occupancy, i.e. sold room nights. Therefore, it relied for the estimation of market shares on the various markets on room capacity, metric that is also widely-used in the industry. For that purpose, Marriott has used the STR database, which provides information on the hotels operating in the EEA, their ownership and room capacity.

(201) Marriott submits that the estimated market shares of the Parties are likely overstated based on that methodology, as the STR database does not include information on all independent hotels operating in the EEA. In addition, Marriott was not able to identify the star-rating of approximately 25% of all hotels in the EEA. These hotels were therefore not taken into account, when estimating the Parties' market shares on markets segmented by star-rating. Moreover, the Parties' serviced apartments and residential offerings have been taken into account for the estimation of Marriott and Starwood's market shares; however, serviced apartments and residential offerings not operated by the Parties were not taken into account for the estimation of the various

187See agreed minutes of a call with a customer of 2 March 2016, a call with a competitor of 11 March 2016, a call with a competitor of 10 March 2016, a call with a competitor of 14 March 2016, responses to Q1a – Questionnaire to competitors, question 19, Q2 – Questionnaire to corporate customers, question 15, Q5 – Questionnaire to travel agencies and tour operators, question 14.

188Competitor sets of the Parties' hotels, as extracted from their revenue optimisation tools and annexed to their management or franchise contracts; Form CO, paragraph 865 and following and Annexes 8.1.a.1, 8.1.a.2 and 8.1.b.1.

189Form CO, paragraphs 324-326.

190Form CO, paragraph 327 and following.

(202) Moreover, Marriott has included in its estimation of market shares all known pipeline hotels scheduled to open until 30 June 2016.

(203) In light of the above, the Commission concludes that the market data provided by Marriott suffice for assessing the impact of the Transaction on the EEA markets for hotel accommodation services. The inclusion of the Parties' pipeline managed hotels planned to start operating beyond that date would not change the assessment of the Transaction, as it would not lead to any additional affected markets and would not affect the conclusion as to the impact of the Transaction on those markets that are affected.

5.2.2 Introduction on affected markets

(204) There are approximately 3.9 million hotel rooms in the EEA according to the STR database. The 5 largest hotel companies on the basis of rooms provide about 15% of the total number of rooms in the EEA, even if the hotels of franchisees were attributed to the brand-parent. The largest hotel company in Europe, Accor, has just [5-10]% share of total hotel rooms, Best Western and IHG being the next largest with a share of just [0-5]%. Hotels operating under Marriott brands (with approximately […] rooms) and hotels operating under Starwood brands (with approximately […] rooms) supply just [0-5]% and [0-5]% of the rooms in the EEA, respectively.

(205) Moreover, independent hotel operators have a strong presence in hotel accommodation services in the EEA, with a room market share of approximately 68%. Using the STR database, approximately 66% of both the 5-star hotels and the 4-star hotels in the EEA were operated as independent hotels in 2015.

(206)

Table 1 below provides an overview of the Parties' individual and combined shares in the hotel accommodation services in each affected market at city and airport level.

5.2.3.1 Closeness of competition

(210) In evaluating the likelihood of non-coordinated effects potentially caused by a concentration, it is important to assess to which extent the products of one merging party are close substitutes to the products sold by the other merging party. The merging firms' incentive to raise prices is more likely to be constrained when rival firms produce close substitutes to the products of the merging firms than when they offer less close substitutes.

5.2.3.1.1 Brand positioning

(211) Both the Parties own hotel brands which are present across the EEA and the majority of competitors which responded to the market investigation questionnaires highlight how the brand of the hotel is very important or somewhat important for the choices of the final customer.

(212) A common element in assessing the effects of the Transaction is therefore the closeness of competition between the brands of the Parties.

(213) To investigate these aspects the Commission assessed whether the hotel brands of the Parties were positioned particularly close to each other.

(214) First, the Parties note that in the ordinary course of business they do not produce documents comparing their respective brands relative to each other or in relation to competing brands. However, in some Transaction-specific documents as well as some third party reports, the Parties seem however to group hotel brands in categories/classes and to compare their brands with a number of competing brands owned by chain operators, as for instance in Figure 1 below.

Figure 1 - Brand positioning map

[…]

Source: Form CO, Annex 7.1.a.3 ("Solar" is Starwood)

(215) From such comparison, it seems that the Parties consider a number of other brands owned by chain operators as being their main competitors in each of the categories/classes identified. In this regard, the Commission notices that most of the international hotel chains have developed a brand portfolio covering several segments and therefore there are several competing brands in each market segment.

(216) Looking at the data submitted by the Parties, it is apparent that Starwood has the majority of its hotels in the EEA under brands which are in the Luxury and Upper Upscale categories/classes, while Marriott has its EEA hotels rather equally spread between in the Luxury, Upper Upscale and Upscale categories/classes. The assessment on closeness of brand positioning will therefore focus on the Luxury and Upper Upscale segment where the Parties' activities overlap.

(217) A number of respondents to the market investigation indicate indeed that they own brands positioned in the Luxury or Upper Upscale category/class, such as:

(a) Accor with its brands Sofitel, Pullman and MGallery;

(b) Carlson with its brands Quorvus Collection and Radisson Blu;

(c) Hilton with its brands Waldorf Astoria, Conrad and Hilton;

(d) Hyatt with its brands Park Hyatt, Grand Hyatt and ANdAZ;

(e) IHG with its brands InterContinental and Crowne Plaza.

(g) Primary competitors of the Tribute Portfolio brand are [Marriott brand, Hilton brand].

200Form CO, Annexes 6.1.1.a.1 and 6.1.1.b.1.

201See responses Q1a – Questionnaire to competitors, question 3.

202See for instance Annex 5.4.13 to the Form CO.

(h) Primary competitors of the Four Points by Sheraton brand are [Hilton brand, Marriott brand, InterContinental brand];

(i) Primary competitors of the Aloft brand are [InterContinental brand, Marriott brand], Citizen, [Hyatt brand];

(219) In addition, data provided by the Parties seem to support the thesis that the Parties' brands are not positioned particularly close to each other. Indeed, between 2013 and 2015, 30 hotels under the Parties' brand switched to a different brand and only one was a switch between the Parties' brands; all the others were hotels which took over another brand whether linked to a chain or an independent one.

(220) In conclusion, considering all evidence available to it, the Commission considers that the Parties' brands in the EEA are not particularly close to each other.

5.2.3.1.2 Geographical proximity

(221) Another common element in assessing the closeness of competition between the Parties is the geographical proximity of their hotels.

(222) The majority of competitors which responded to the market investigation highlight how the location of the hotel is a very important factor in the choices of the final customer.

(223) Indeed if the Parties' hotels were particularly close to each other in geographical terms, the competition lost with the Transaction might be higher than if the hotels were not particularly close to each other.

(224) The Commission, in its assessment of some cities where the combined market shares of the Parties are the highest, included an assessment of whether the Parties' hotels were geographically particularly close to each other or whether competing hotels could be found at similar or shorter distances.

5.2.3.2 Consumer brand loyalty and loyalty programs

5.2.3.2.1 Marriott's views

(225) Marriott submits that the market in Europe is very fragmented, the EU consumer is generally brand agnostic and has a low understanding of the different loyalty programs value propositions. Consequently, according to Marriott, loyalty programs have a very limited penetration in the EU marketplace and the Transaction is not going to change the competitive landscape with regards to the Parties' loyalty programs.

(226) According to the respondents to the market investigation, the loyalty programs do not seem to be a strong driver of consumer decisions in the EEA and were ranked as the lowest or second lowest criteria out of seven criteria taken into account by consumers when choosing a hotel.

(227) The Parties and at least one competitor, which was contacted during the market investigation, appear to use loyalty programs to collect data about customers in order to provide more targeted services and experiences to the final customers.

(228) Contacts with market participants highlighted also the fact that a single customer can be a member of several competing hotel loyalty programs.

(229) Moreover, third-party reports purchased by the Parties for their internal use show how European customers are not particularly brand loyal and do not use the loyalty programs as main driver for their choices. For instance, the fact that a hotel is part of a frequent guest programs is ranked 84 in order of importance by EU consumers, while it is ranked 14 by consumers in the USA. Moreover, the following quotes can be found: "The general lack of brand preference and loyalty is reflected in European travellers' unwillingness to pay extra to stay at their preferred hotel" and "Across the globe price and location are key criteria in hotel selection, but European consumers are very focused on these dominant pragmatic factors – for both leisure and business travel".

5.2.3.2.3 Conclusion

(230) In light of all evidence available to it, the Commission considers that brand loyalty of consumers, as well as loyalty programs, does not play any material role in the competitive assessment of the Transaction in the markets for hotel accommodation services in the EEA.

208Form CO, paragraph 925.

209See responses Q1a – Questionnaire to competitors, question 5; Q1b – Questionnaire to independent and franchised competitors, question 4 and Q5 – Questionnaire to travel agencies and tour operators, question 4.

210Form CO, paragraph 169. See also agreed minutes of a call with a competitor on 14 March 2016.

See agreed minutes of a call with a market participant on 14 March 2016 and agreed minutes of a call with a market participant on 16 March 2016. Marriott also mentions that "typically, corporate/business travellers (being the principal customer base for loyalty programs), will, similarly to their leisure counterparts, be brand agnostic and as such, hold multiple memberships to hotel loyalty programs" (Form CO, paragraph 175).

212Annex 8.2.a.2 to the Form CO.

5.2.3.3 Barriers to entry or expansion

(231) The Commission is unlikely to find that the merger will create or strengthen a dominant position or otherwise significantly impede effective competition when rival firms have available capacity and find it profitable to expand output sufficiently. In other words, the extent to which competitors to the merged entity constrain the merged entity from raising prices not only depends on the level of their spare capacity but also on whether these firms have the incentive to react aggressively to a post-merger price increase.

(232) Some proposed mergers would, if allowed to proceed, significantly impede effective competition by leaving the merged firm in a position where it would have the ability and incentive to make the expansion of smaller firms and potential competitors more difficult or otherwise restrict the ability of rival firms to compete.

5.2.3.3.1 Marriott's views

(233) Marriott puts forward that there are no material legal or regulatory barriers to prevent market entry and more specifically, there are no intellectual property rights or other related rights that would prevent market entry and that will remain the case post-Transaction.

(234) In the frame of an asset-light business model, entry can take place on the basis of management or franchise agreements with property owners. This implies that building costs, site availability and issuance of planning permissions are not applicable to hotels operators which might want to enter a specific market but rather the only barriers for a potential new entrant are the possession of a brand and experience in managing a hotel business is important for entry.

(235) Marriott cites examples of existing hotel operators which entered several markets in the EEA such as Rosewood, Shangri-La, Mandarin Oriental and the Peninsula hotel group.

5.2.3.3.2 Commission's assessment

(236) The Commission first considers that not all competitors operate with an asset-light model and therefore some regulatory barriers related to the actual estate of the hotel can apply to certain competitors.

(237) Second, there is a degree of dynamism in the market with hotels changing brands and thus giving possibilities to competitors to enter the market.

(243) Therefore, post-Transaction, the market shares of the merged entity would remain moderate.

(244) In Vienna, a number of other 5-star chain hotels compete actively in the market, including international chain operators such as Accor, Hyatt, IHG, Carlson, Kempinski and Steigenberger.

(245) Moreover, additional competition stems from independent 5-star hotels, like The Guesthouse Vienna and Hotel Sans Souci Vienna. According to customers' reviews on TripAdvisor, the 1 and 2 best 5-star hotels in Vienna are independent hotels (The Guesthouse Vienna and Hotel Sans Souci Vienna respectively), while the Parties' 5-star hotels are ranked 15, 21, 29, 77 and 96. Therefore, the competitive constraint posed by 5-star independent hotel appears to be strong vis-à-vis the 5-star hotels of the Parties.

Closeness of competition

(246) First, as described in Section 5.2.3.1.1, the Commission considers that the Parties' brands are not particularly close to each other.

(247) An important part of the assessment of closeness of competition is the analysis of the competing hotels which each of Marriott's and Starwood's hotels monitors as their point of reference in the competitive landscape. That analysis focuses on which nearby hotels are included in the monitoring activities by each single hotel in order to set its room prices, which, as the market investigation results confirm, are one of the main drivers for the choice of the final customers and ultimately of the hotel success.

(248) Each of the Marriott and Starwood hotels uses a yield management software which takes into account a number of variables (among others the occupancy rates and average daily rate) from a peer group of competitive hotels selected by hotel management, the so-called "competitive set". Each hotel in the competitive set has a specific weight in the benchmarking.

(249) Each of the Marriott 5-star hotels has in its competitive set only some of the Starwood 5-star hotels, together with other 5-star chain hotels; one Marriott hotel includes in its competitive set also one independent hotel while another Marriott 5-star hotel includes a 4-star hotel in its competitive set. The relative weight given to the Starwood hotels within the competitive set of each of the Marriott hotels is equal or lower to the weight given to the other hotels in the competitive set for the majority of the year.

(250) Each of the Starwood 5-star hotels has in its competitive set only one of the Marriott 5-star hotels, together with other 5-star chain hotels; two Starwood hotels include in their competitive sets also one independent hotel and a 4-star hotel. The relative weight given to the Marriott hotels within the competitive set of each of the Starwood hotels is equal to or lower than the weight given to the other hotels in the competitive set.

(251) Therefore, Marriott 5-star hotels do not consider Starwood 5-star hotels particularly close competitors and vice versa.

(252) When considering services and amenities, all of the Parties' 5-star hotels and competing 5-star hotels analysed by the Parties offer 24h front desk, a gym and a restaurant.

(253) All of Marriott hotels have a swimming pool, whereas only one of the Starwood hotels has a swimming pool. Further, the hotels of Accor, Hyatt, Kempinski and a number of independent hotels also offer a swimming pool to their guests.

(254) Two of the Starwood hotels and one Marriott hotel also offer a spa, as do the hotels of Accor, Hyatt, IHG, Kempinski, Steigenberger and several independent hotels.

(255) When considering prices to customers, the Parties do not seem to be particularly close competitors either: the Parties face competition, from both chain hotels and independent hotels, at the various price points that they apply in their 5-star hotels.

(256) Therefore, the Parties' hotels are not particularly close competitors with regards to amenities, services and prices offered to their customers.

(257) On a geographical level, when considering the surroundings of the Parties' hotels, it can be noticed that, while some of the Parties' 5-star hotels are close to each other, others are more distant. When including competitor's 5-star hotels, it becomes apparent that the Parties' hotels are not particularly close to each, as shown in Figure 2. Indeed, there are a number of 5-star independent hotels and one under an IHG brand in the same area.

225 The relative weight can be changed on a daily basis; the assessment in this Decision is based on monthly overview as of November 2015. See Table 15 of the Form CO.

226 Form CO, Table 13.

227 Form CO, Table 13.

228 Form CO, Table 13.

229 Average price range double room (STR).

230 Form CO, Table 13. For example the Marriott Ritz Carlton Vienna hotel offers rooms at a price range of EUR 449 – 799, similarly the Starwood Hotel Imperial offers rooms at a price range of EUR 539 – 606; among competing 5-star hotels the Park Hyatt Vienna hotel offers rooms at a price range of EUR 510 – 674, the Grand Hotel Wien offers rooms at a price range of EUR 508 – 640, the Hotel Sacher Wien offers rooms at a price range of EUR 570 – 881, and the Ring Hotel offers rooms at a price range of EUR 558 – 636.

Figure 2 - Competitive map of Vienna

(258) Therefore, the Parties' hotels are not particularly close competitors in Vienna on a geographical level.

(259) In the light of the above, the Commission concludes that Marriott and Starwood are not particularly close competitors in Vienna in the market for hotel accommodation services or in any plausible further segmentation of such market.

Other elements

(260) The majority of all the respondents to the market investigation indicate that, in the market for hotel accommodation services in Vienna, there will be sufficient competition to prevent the merged entity from raising prices post-Transaction.

231 See responses Q1a – Questionnaire to competitors, question 66; Q1b – Questionnaire to independent and franchised competitors, question 50; Q2 – Questionnaire to corporate customers, question 22; Q3 – Questionnaire to consumer associations, question 17 and Q5 – Questionnaire to travel agencies and tour operators, question 29.

(266) Therefore, post-Transaction the market shares of the merged entity would remain moderate.

(267) In Milan, a number of 5-star chain competitors will remain active in the market, such as Grupo Sol Melia, Dorchester Collection, Four Seasons, Hyatt and Mandarin Oriental.

Closeness of competition

(268) First, as described in Section 5.2.3.1.1, the Commission considers that the Parties' brands are not particularly close to each other.

(269) Moreover, when considering prices to customers, the Parties do not seem to be particularly close competitors: the Parties face competition from other chain hotels at the various price points that they apply in their 5-star hotels.

(270) When considering services and amenities, what the Parties' managed 5-star hotels offer to their customers is not particularly different from what competing 5-star hotels analysed by the Parties offer to their customers.

(271) In the light of the above, the Commission concludes that Marriott and Starwood are not particularly close competitors in Milan in the market for hotel accommodation services or in any plausible further segmentation of such market.

Other elements

(272) The majority of all the respondents to the market investigation indicate that, in the market for hotel accommodation services in Milan, there will be sufficient competition to prevent the merged entity from raising prices post-Transaction.

5.2.4.2.3 Conclusion

(273) Overall, the Commission concludes, on the basis of, in particular, the moderate combined market shares of the Parties, the presence of strong competitors, and the fact that a majority of market participants indicates that there would be sufficient competitive pressure post-Transaction, that the Transaction would not lead to any significant competition concerns in the markets for hotel accommodation services in Milan.

235 Form CO, Table 24.

236 Average price range double room (STR).

237 Form CO, Table 24. For example the Marriott Bulgari Hotel Milano offers rooms at a price range of EUR 710 – 764, no Starwood hotel offers rooms at a similar price range while, among competing 5-star hotels, the Mandarin Oriental Milan hotel offers rooms at a price range of EUR 769 – 1044, the Four Seasons hotel Milano offers rooms at a price range of EUR 718 – 1006, and the Armani Hotel Milano hotel offers rooms at a price range of EUR 676 – 811.

238 Form CO, Table 24.

239 See responses Q1a – Questionnaire to competitors, question 66; Q1b – Questionnaire to independent and franchised competitors, question 50; Q2 – Questionnaire to corporate customers, question 22; Q3 – Questionnaire to consumer associations, question 17 and Q5 – Questionnaire to travel agencies and tour operators, question 29.

(281) First, as described in Section 5.2.3.1.1, the Commission considers that the Parties' brands are not particularly close to each other.

(282) An important part of the assessment of closeness of competition is the analysis of the competing hotels which each of Marriott's and Starwood's hotels monitors as their point of reference in the competitive landscape. That analysis focuses on which nearby hotels are included in the monitoring activities by each single hotel in order to set its room prices, which, as the market investigation results confirm, are one of the main drivers for the choice of the final customers and ultimately of the hotel success.

(283) On one hand, the competitive sets of the 5-star hotels managed by Marriott include only two of the 5-star hotels of Starwood, together with other 5-star chain hotels, 5-star independent hotels and one 4-star hotel. The relative weight given to the Starwood hotels in the competitive sets is lower than the weight given to the other hotels in the competitive set for the majority of the year.

(284) On the other hand, Starwood hotels do not have any Marriott hotel in their competitive sets but have a number of 5-star chain hotels, 5-star independent hotels and sporadically 4-star hotels.

(285) Therefore, Marriott 5-star hotels do not consider Starwood 5-star hotels particularly close competitors, and vice versa.

(286) When considering services and amenities, all of the Parties' 5-star hotels and competing 5-star hotels analysed by the Parties offer 24h front desk and a restaurant.

(287) All of Marriott hotels have a spa, whereas only one of the Starwood hotels has a spa; further, the hotels of Hilton, Kempinski and most of independent hotels offer a spa.

(288) One Marriott hotel offers a swimming pool whereas no Starwood hotel has one; further, the hotels of Hilton, Kempinski and one independent hotel offer a swimming pool.

(289) One Marriott hotel offers a gym, whereas two Starwood hotels have a gym; further, the hotels of Hilton, Kempinski and most of the independent hotels offer a gym.

(290) When considering prices to customers, the Parties do not seem to be particularly close competitors either: the Parties face competition, from both chain hotels and independent, at the various price points that they apply in their 5-star hotels, with the only exception of the Westin Europa & Regina Venice which has price ranges above all of the competitors.

(291) Therefore, the Parties' hotels are not particularly close competitors with regards to amenities, services and prices offered to their customers.

(292) On a geographical level, when considering the surroundings of the Parties' hotels, it can be noticed that the Parties' 5-star hotels are not particularly close to each other. When including competitor's 5-star hotels it becomes even more apparent that the Parties' hotels are not particularly close to each other, as shown in Figure 3. Indeed, there are one 5-star chain hotel under a Hilton brand, another 5-star chain hotel under a Kempinski brand and a number of 5-star independent hotels in the same area.

248 Form CO, Table 28.

249 Form CO, Table 28.

250 Average price range double room (STR).

251 Form CO, Table 28. For example the JW Marriott Venice Resort & Spa offers rooms at a price range of EUR 344 – 1132, the Starwood Luxury Collection Hotel Danieli offers rooms at a price range between EUR 540 – 1128; among competing 5-star hotels, the Belmond Hotel Cipriani Venice offers rooms at a price range of EUR 661 – 1127, the San Clemente Palace Kempinski Venice offers rooms at a price range of EUR 680 – 1266, the Hotel Excelsior Venice Lido Resort offers rooms at a price range of EUR 550 – 1012, and the Palazzina G offers rooms at a price range of EUR 459 – 1108.

Figure 3 – Competitive map of Venice

(293) Therefore, the Parties' hotels are not particularly close competitors in Venice on a geographical level.

(294) In the light of the above, the Commission concludes that Marriott and Starwood are not particularly close competitors in Venice in the market for hotel accommodation services or in any plausible further segmentation of such market.

Other elements

(295) The majority of all the respondents to the market investigation indicate that in the market for hotel accommodation services in Venice there will be sufficient competition to prevent the merged entity from raising prices post-Transaction.

252 See responses Q1a – Questionnaire to competitors, question 66; Q1b – Questionnaire to independent and franchised competitors, question 50; Q2 – Questionnaire to corporate customers, question 22; Q3 – Questionnaire to consumer associations, question 17 and Q5 – Questionnaire to travel agencies and tour operators, question 29.

(301) Therefore, post-Transaction the market shares of the merged entity would remain moderate.

(302) In Amsterdam a number of 5-star chain competitors will remain active in the market, such as Accor, Carlson, Hyatt, IHG, NH and Hilton.

Closeness of competition

(303) First, as described in Section 5.2.3.1.1, the Commission considers that the Parties' brands are not particularly close to each other.

(304) When considering prices to customers and services and amenities, the Parties do not seem to be particularly close competitors either: the Parties face competition from other chain hotels at the various price points that they apply in their 5-star hotels and offer similar services and amenities.

(305) In the light of the above, the Commission concludes that Marriott and Starwood are not particularly close competitors in Amsterdam in the market for hotel accommodation services or in any plausible further segmentation of such market.

Other elements

(306) None of the respondents to the market investigation considers that, in the market for hotel accommodation services in Amsterdam, there will not be sufficient competition to prevent the merged entity from raising prices post-Transaction.

5.2.4.4.3 Conclusion

(307) Overall the Commission concludes, on the basis of, in particular the moderate combined market shares of the Parties, the presence of strong competitors, and the absence of concerns expressed by market participants, that the Transaction would not lead to any significant competition concerns in the hotel accommodation markets in Amsterdam.

(308) In view of the above, and considering all evidence available to the Commission, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in respect of the markets for hotel accommodation services in Amsterdam, under any plausible segmentation.

256 Form CO, Table 31.

257 Average price range double room (STR).

258 Source: Form CO, Table 31. Prices for the W Hotel Amsterdam are not yet included in STR due to its recent opening. However, for example the Amsterdam Marriott Hotel offers rooms at a price range of EUR 261 – 295; among competing 5-star hotels, the Hilton Amsterdam offers rooms at a price range of EUR 255 – 377, the Hotel Okura Amsterdam offers rooms at a price range of EUR 235 – 403 and the Westmont Bilderberg Garden Hotel offers rooms at a price range of EUR 210 – 249.

259 See responses Q1a – Questionnaire to competitors, questions 70 and 71; Q1b ¬ Questionnaire to independent and franchised competitors, questions 54 and 55; Q2 – Questionnaire to corporate customers, questions 24 and 25; Q3 – Questionnaire to consumer associations, questions 19 and 20 and Q5 – Questionnaire to travel agencies and tour operators, questions 31 and 32.

(313) When considering only 5-star chain hotels in Warsaw, a number of strong competitors are active and will continue to compete with the merged entity also post-Transaction. IHG and Accor have market shares of respectively [10-20]% and [10-20]%, whereas other chains, such as Carlson and Mamaison are also present. Moreover, additional material competition stems from independent 5-star hotels like H15 Boutique, Rialto, Bellotto, and Regent Warsaw Hotel. According to customers' reviews on TripAdvisor, the 1st and 2nd best 5-star hotels in Warsaw are independent hotels (H15 Boutique and Rialto respectively), another 5-star independent hotel is ranked 6th while the Parties' 5-star hotels are ranked 8, 10, 15 and 16. Therefore, the competitive constraint posed by 5-star independent hotel appears to be strong vis-à-vis the 5-star hotels of the Parties.

(314) Moreover, in addition to existing competing 5-star hotels, some respondents to the market investigation explain that a number of new 5-star hotels are planned to open in Warsaw in the coming years. It is expected that, independently of the Transaction, an important number of new rooms in competing 5-star chain hotels will become available in a timely manner (in the course of 2017). Marriott estimates that, when considering such pipeline projects, the Parties’ combined share for 5-star hotels in Warsaw would decrease to [40-50]%.

Closeness of competition

(315) First, as described in Section 5.2.3.1.1, the Commission considers that the Parties' brands are not particularly close to each other.

(316) In the specific case of Warsaw, the Marriott hotel is under the Marriott brand, which is an Upper Upscale brand (See Figure 1 above) and only two of the three Starwood hotels are under Upper Upscale brands (i.e. Westin and Sheraton) whereas the third Starwood hotel is under a Luxury brand, (i.e. The Luxury Collection). At the same time, the competing hotels are also spread between the two brand categories/classes, with IHG having its 5-star hotel under the Intercontinental brand and Accor having its hotel under the Sofitel brand, which are both classed as Luxury, whereas Carlson has its 5-star hotel under the Radisson Blu brand, which is classed as Upper Upscale. Therefore, the Parties' brands are not particularly close to each other in the market for 5-star hotels in Warsaw.

(317) Second, an important part of the assessment of closeness of competition is the analysis of the competing hotels, which each of Marriott's and Starwood's hotels monitor as their point of reference in the competitive landscape. This analysis focuses on which nearby hotels are monitored, notably through hotels' revenue management system in order to set its room prices, which, as derives from the market investigation, is a key competition parameter.

(318) The Marriott 5-star hotel has in its competitive set two of the Starwood 5-star hotels, together with other 5-star chain hotels and one 4-star hotel. The relative weight given to the Starwood hotels within the competitive set of the Marriott 5-star hotel is however much lower than the weight given to the other hotels in the competitive set. More concretely, the Starwood hotels are given a relative weight of […] during weekend nights and […] during weekday nights, whereas the other hotels in the set are given relative weights of […] during weekend nights and […] during weekday nights. Hence, the Marriott hotel considers hotels not operating under the Starwood brands as its closest competitors in Warsaw.

(319) One of the Starwood hotels has in its competitive set the Marriott 5-star hotel, together with one 4-star chain hotel and one 3-star chain hotel; the second Starwood hotel has in its competitive set the Marriott 5-star hotel together with other 5-star hotels, one of which is an independent hotel; the third Starwood hotel does not have the Marriott 5-star hotels in its competitive set but it has only other 5-star chain hotels.

(320) Therefore, the assessment of the monitoring practices and of the competitive sets of the relevant hotels demonstrates clearly that for the Marriott 5-star hotel, the Starwood 5-star hotels are not particularly close competitors, and vice versa.

(321) Third, when considering the price rates of 5-star hotels in Warsaw, the Parties do not appear to be particularly close competitors either. The price range of the Marriott 5-star hotel is EUR 205-302, whereas the price ranges offered by the three Starwood hotels are EUR 71-160, EUR 144-244, and EUR 167-262. Overall the Parties face competition, from both chain and independent hotels, at the various price points that they apply in their 5-star hotels.

(322) When considering services and amenities, all of the Parties' 5-star hotels and competing 5-star hotels analysed by the Parties offer 24h front desk, a spa, a gym and a restaurant.

(323) The Marriott 5-star hotel and one of the Starwood 5-star hotels have a swimming pool, just like the 5-star hotels of Accor, Carlson, IHG, Mamaison and one of the independent 5-star hotels.

(324) Therefore, the Parties' hotels are not particularly close competitors with regard to amenities, services and prices offered to their customers.

(325) Fourth, on a geographical level, when considering the surroundings of the Parties' hotels it can be noticed that the Parties' 5-star hotels are not particularly close to each other. When including competitor's 5-star hotels it becomes even more apparent that the Parties' hotels are not particularly close to each other, as shown in Figure 4, indeed there are in the same area one 5-star chain hotel under a IHG brand, one 5-star chain hotel under an Accor brand and another 5-star chain hotel under a Carlson brand and a number of 5-star independent hotels.

Figure 4 – Competitive map of Warsaw

Therefore, the Parties' hotels are not particularly close competitors in Warsaw on a geographical level.

(326) In the light of the above, the Commission concludes that Marriott and Starwood are not particularly close competitors in Warsaw in the market for hotel accommodation services or in any plausible further segmentation of such market.

Other elements

(327) The large majority of all the respondents to the market investigation indicate that in the market for hotel accommodation services in Warsaw there would be sufficient competition to prevent the merged entity from raising prices post-Transaction. For instance, an respondent to Q5 – Questionnaire to travel agencies and tour operators claims that "[Warsaw] offer[s] a variety of properties/competitors with same or similar standards than Marriott and Starwood"; another respondent to Q2 – Questionnaire to corporate customers points out that "Sufficient alternatives in the market plus the hotels compete with each other anyway, based on different ownership".

(328) It is also relevant to recall, that, as mentioned in Section 5.2.3.2, consumers in Europe are not particularly brand loyal and are less willing to pay a price premium to stay at their preferred brand hotels. As the 5-star hotels of the Parties are not particularly close geographically, a number of other 5-star hotels operate in the city and customers' ability to easily access information on all hotels' pricing and service offerings, notably through OTAs, it is likely that any price increase by the Parties will see a decline in the occupancy of their rooms.

5.2.4.5.3 Conclusion

(329) Overall the Commission concludes, in particular on the basis of the presence of strong chain competitors, together with the high number of material independent competitors, the likely and timely entry of new competitors, the absence of particular closeness of competition between the Parties, and the fact that a large majority of market participants indicates that there would be sufficient competitive pressure post-Transaction, that the Transaction would not lead to any significant competition concerns in the markets for hotel accommodation services in Warsaw.

(330) In view of the above, and considering all evidence available to the Commission, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in respect of the markets for hotel accommodation services in Warsaw, under any plausible segmentation.

5.2.4.6 Barcelona

5.2.4.6.1 Parties' activities

(331) There are approximately 41 659 hotel rooms in total in Barcelona. Of the 41 659 hotel rooms in Barcelona, 20 510 rooms are in 4-star and 5 137 in 5-star hotels.

273 See responses Q1a – Questionnaire to competitors, question 66; Q1b ¬ Questionnaire to independent and franchised competitors, question 50; Q2 – Questionnaire to corporate customers, question 22; Q3 – Questionnaire to consumer associations, question 17 and Q5 – Questionnaire to travel agencies and tour operators, question 29.

274 Q5 – Questionnaire to travel agencies and tour operators, question 29.

275 Q2 – Questionnaire to corporate customers, question 22.

72

Closeness of competition

(337) First, as described in Section 5.2.3.1.1, the Commission considers that the Parties' brands are not particularly close to each other.

279 (338) When considering prices to customers and services and amenities, the Parties do not seem to be particularly close competitors: the Parties face competition, from both chain hotels and independent, at the various price points that they apply in their 5-star hotels and offer similar services and amenities.

In addition, on a geographical level, when considering the surroundings of the Parties' hotels it can be noticed that the Parties' 5-star hotels are not particularly close to each other. When including competitors' 5-star hotels, it becomes even more apparent that the Parties' hotels are not particularly close to each other, as shown in Figure 5, indeed there are in the same area a number of 5-star chain hotel under brands from Accor, Eurostars, Derby Hotels Collection and Mandarin Oriental as well as a number of 5-star independent hotels. Figure 5 – Competitive map of Barcelona

279 Average price range double room (STR).

280 Form CO, Table 40. For example the Marriott Renaissance Barcelona Hotel offers rooms at a price range of EUR 238 – 372, the Starwood Le Méridien Barcelona offers rooms at a price range of EUR 270 – 338; among competing 5-star hotels, the Pullman Barcelona Skipper hotels offers rooms at a price range of EUR 204 – 421, the Eurostars Hotel Grand Marina offers rooms at a price range of EUR 286 – 327, the Melia Barcelona Sarria offers rooms at a price range of EUR 199 – 309, the Grand Hotel Central offers rooms at a price range of EUR 239 – 331 and the Husa Hotel Miramar Barcelona offers rooms at a price range of EUR 243 – 388.

74

(339) Therefore, the Parties' hotels are not particularly close competitors in Barcelona on a geographical level.

(340) In the light of the above, the Commission concludes that Marriott and Starwood are not particularly close competitors in Barcelona in the market for hotel accommodation services or in any plausible further segmentation of such market.

Other elements

(341) The majority of all the respondents to the market investigation indicate that in the market for hotel accommodation services in Barcelona there would be sufficient competition to prevent the merged entity from raising prices post-Transaction.

5.2.4.6.3 Conclusion

(342) Overall, the Commission concludes, on the basis of, in particular the moderate combined market shares of the Parties, the high number of independent competitors together with the presence of strong chain competitors, the absence of specific closeness of competition between the Parties, and the fact that a majority of market participants indicates that there would be sufficient competitive pressure post-Transaction, that the Transaction would not lead to any significant competition concerns in the markets for hotel accommodation services in Barcelona.

(343) In view of the above, and considering all evidence available to the Commission, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in respect of the markets for hotel accommodation services in Barcelona, under any plausible segmentation.

5.2.4.7 London Heathrow

5.2.4.7.1 Parties' activities

(344) There are approximately 12 127 hotel rooms in total in the London Heathrow area. Of these, 7 782 rooms are in 4-star and 605 in 5-star hotels.

(345) Both Parties operate 4-star hotels in the London Heathrow airport area: Marriott manages [names of hotels]. Starwood manages [names of hotels].

(346) The Parties' activities overlap in the London Heathrow airport area as well as in the city of London; the Parties' activities do not overlap in managing hotels in any other airport of London.

5.2.4.7.2 Commission's assessment

Market shares and competitors

(347) When considering all plausible markets for hotel accommodation services in the London Heathrow airport area, the markets for 4-star hotels would be affected with a combined market share of the Parties of [20-30]%:

282 Moreover, the 4-star [name of hotel] is operated under a Marriott brand in the London Heathrow airport area; however, as analysed in Section 5.2.1.2, this hotel is not controlled by Marriott, but instead operated by their respective hotel owner and/or manager and therefore in competition with the Parties' hotels in the London Heathrow airport area.

283 The Marriott-branded 4-star hotels are [names of hotels], and the 5-star hotels are [names of hotels]. For completeness, Marriott also franchises its brand name to [...], which is a timeshare business. The Starwood-branded 4-star hotels are [names of hotels], and the 5-star hotels are [names of hotels] (Parties' answer of 30 May 2016 to RFI 6).

284 The Parties do not overlap in managing airport hotels in any other London airport and when considering the city of London, their combined market share does not reach 20% under any plausible market definition. Following a worst-case approach, the market shares presented are on a London Heathrow-only basis and do not include other airport hotels nor city hotels.

285 When considering only chain hotels, the Parties would have [20-30]% of all the 4-star chain hotel rooms and [20-30]% of all the 4- and 5-star hotel chain rooms.

76

5.2.4.7.3 Conclusion

(354) Overall the Commission concludes, on the basis of, in particular, the moderate combined market shares of the Parties, the presence of strong competitors, and the absence of concerns by market participants, that the Transaction would not lead to any significant competition concerns in the markets for hotel accommodation services in the London Heathrow airport area.

(355) In view of the above, and considering all evidence available to the Commission, the Commission considers that the Transaction does not raise serious doubts as to its compatibility with the internal market in respect of the markets for hotel accommodation services in the London Heathrow airport area, under any plausible segmentation.

5.3 Hotel management services

(356) As indicated in Sections 4.2.2 and 4.3.2, the Commission will assess the effects of the Transaction on the plausible market of the provision of hotel management services by hotel chains in the EEA and worldwide.

2905.3.1 Methodology for estimating market shares

(357) Throughout the EEA, in 2015, Marriott provided management services to […] hotels owned by third parties for a total of […] rooms and Starwood to […] hotels owned by third parties for a total of […] rooms. Based on all hotel rooms across all star categories in the EEA, Marriott and Starwood managed together approximately [0-25]% of the total stock of rooms in the EEA.

(358) At worldwide level, in 2015, Marriott provided management services to around 1 200 hotels owned by third parties for a total of 314 000 rooms and Starwood to around 600 hotels owned by third parties for a total of 195 000 rooms. Based on all

290 In line with the approach taken for the provision of hotel accommodation services, Marriott estimates the market shares of the Parties and of their competitors for the provision of hotel management services on the basis of the number of managed rooms. Based on its investigation, the Commission agrees that the total number of rooms managed for third-party owners (total capacity in volume) is a fair proxy of the total size of the market and that using the number of managed rooms to calculate market shares is the most robust model in the context of this Decision.

291 There is a marginal discrepancy with the data set out in Section 1 due to the difference in the reference period.

292 Form CO, paragraph 266 and Annex 6.3.3. Marriott has calculated those market shares on the basis of the STR database, according to which Marriott managed […] rooms and Starwood managed […] rooms in the EEA in 2015 out of 3.94 million total hotel rooms in the EEA. Those data deviate from the actual data provided by the Parties and presented in this Section 5.3.1 of this Decision, an inaccuracy which will be further discussed in the next paragraphs of this section.

78

293 hotel rooms across all star categories worldwide, Marriott and Starwood managed together approximately 14.8% of the total stock of rooms worldwide.

(359) In order to translate those data into the Parties' position on the market for the provision of hotel management services to third-party owners at EEA and worldwide levels, a distinction must be made, within the total stock of rooms, between rooms of hotels that are self-managed by the hotel owner (not to be taken into account to estimate market shares) and rooms of hotels that are managed by a third-party management company (to be taken into account to estimate market shares).

Exclusion of hotels self-managed by their owner

(360) For that purpose, Marriott proposes two approaches for hotels in the EEA:

a. based on ownership information in STR, by excluding the hotels registered in the STR database as owned and managed by the same company; and

b. based on publicly available information on ownership, by excluding the number of hotel rooms deemed owned and managed by the same company according to annual reports.

(361) However, Marriott acknowledges that the first approach is flawed due to the incomplete information regarding management arrangements and ownership in STR. In particular, the hotel owner is reported in the STR database in less than 10% of the hotels listed as located in the EEA. Therefore, Marriott admits that it has not been able to exclude comprehensively hotels owned and managed by the same company from the total market for the hotel management services in the EEA.

(362) In addition to the flaws pointed out by Marriott, the Commission notes that the first approach gives rise to materially inaccurate results for the Parties, although the latter own a marginal proportion of the hotels they manage.

(363) The Commission considers, as the Parties do, that the second approach is likely to lead to a more reliable estimation of the Parties' position on the market for the provision of hotel management services at the EEA level than the first approach.

(364) Nevertheless, the Commission expresses reservations as to the extrapolation of the average ownership ratio (36.8%) to small chains. Indeed, the market investigation results provided indications that the shift from hotel ownership to hotel management and franchising is specific to large hotel chains that have reached a certain level of maturity and strength in the hotel industry. In addition, each Best Western branded hotel is independently owned and operated; therefore the Commission takes the view that Best Western, among others, does not provide hotel management services. Therefore, the approach adopted by Marriott is likely to lead, for smaller chains, to an overestimation of the total number of hotels they manage for third-party owners and to an underestimation of the Parties' market share for the management services.

(365) Finally, Marriott submits that it is not in a position to estimate the number of hotels that are self-managed or managed by third-party management companies worldwide. Marriott indicates that the flaws identified at EEA level are aggravated by the unreliability of the total number of available rooms worldwide, due to the poor coverage of STR outside of the EEA and North America and its inability to have access to data sources other than STR.

(366) Given the limitations of Marriott's approaches to market share determination for the management services by hotel chains provided to third-party owners, the Commission has undertaken a partial market reconstruction exercise at EEA and worldwide levels. The Commission has received information about the number of hotels managed for third-party owners in the EEA and/or worldwide, the corresponding number of rooms and the corresponding management fees received from 25 hotel chains active in the EEA (other than the Parties), among which e.g. Hilton, IHG, Accor, Carlson, Melia, Groupe du Louvre.

(367) That market reconstruction exercise did not aim at obtaining a fully comprehensive view of the market for hotel management services but rather at (i) testing the methodological assumptions applied by Marriott at EEA level in its second approach, (ii) comparing the size of the Parties and of their biggest chain competitors using actual data from those competitors; and (iii) estimating the maximum market shares of the Parties by determining their position on a part of it (i.e. hotel management services provided by the main hotel chains active in the EEA).

298 As an illustration, the numbers of rooms managed by the Parties for third-party owners according to Annex 6.3.2 deviate only by 2%-3% from the actual numbers of rooms that the Parties declare they manage.

299 See agreed minutes of a call with a competitor on 14 March 2016.

300 See www.bestwestern.com.

301 While Best Western would hold a [5-10]% market share according to the Form CO, Annex 6.3.2.

302 See above in this Section 5.3.1 of this Decision.

303 See Marriott's answer of 02 June 2016 to RFI 11.

80

Parties by determining their position on a part of it (i.e. hotel management services provided by the main hotel chains active in the EEA).

5.3.2 Estimated market shares of the Parties and affected markets

5.3.2.1 Marriott's views

(368) Marriott submits that the Parties' combined market share does not exceed 20% at EEA-wide level or at worldwide level.

(369) At EEA-wide level, Marriott estimates that the Parties' combined market share on the market for the provision of hotel management services by hotel chains would be [5-10]% (Marriott: [0-5]%; Starwood: [0-5]%) following the first approach described in Section 5.3.1 of this Decision (based on ownership information in STR) and [5-10]% (Marriott: [5-10]%; Starwood: [0-5]%) following the second approach described above in Section 5.3.1 of this Decision (based on publicly available information on ownership).

(370) Upon further questions of the Commission, Marriott eventually states that it cannot provide reliable market share estimates for the Parties at worldwide level.

5.3.2.2 Results of the market reconstruction

(371) The results of the market reconstruction deviate significantly from Marriott's submission and do not enable to exclude that the market for the provision of hotel management services would be horizontally affected both at EEA level, where the combined market share of the Parties could reach up to [30-40]% (Marriott: [10-20]%; Starwood: [10-20]%), and at worldwide level, where the combined market share of the Parties could reach up to [40-50]% (Marriott: [20-30]%; Starwood: [10-20]%).

(372) Although the market reconstruction exercise is likely to materially overstate the Parties' market share, the Commission considers that the assumption of the Parties with regard to their limited position on the market for the provision of hotel management services cannot be confirmed at this stage.

(373) The Commission will thus further examine the horizontal effects of the Transaction on the market for the provision of hotel management services.

5.3.3 Effects of the Transaction on ongoing management agreements

(374) The Commission considers that the Parties' current customers of hotel management services are a priori sheltered against price increases until termination or expiry of their existing management agreement, considering (i) the long-term duration of management agreements entered into by the Parties (between [...] and [...] years), and

304 Form CO, paragraph 809.

305 Form CO, Annex 6.3.1.

306 See Marriott's answer of 02 June 2016 to RFI 11.

307 Form CO, Tables 4 and 5.

81

(ii) the determination of the financial conditions, in particular the level of management fees, for their entire duration without right to adjustment for the hotel management company.

(375) In this regard, the Commission notes that no owner of hotels managed by the Parties that responded to the market investigation questionnaires raises any pricing concern, when asked about the impact, if any, of the Transaction on their company. In fact, most respondents foresee a positive impact on their owned hotels managed by the Parties, thanks in particular to the extended scope of the loyalty program among the Parties' brands (hence an increased attractiveness of the hotels managed by the Parties) and an extended breadth of brand opportunities (hence an increased ability to switch brands for the hotel owners).

(376) The Commission therefore concludes that the pricing effects of the Transaction, if any, may only materialise at the time of award of a new management agreement (either for a new hotel or following termination or expiry of an existing management agreement). The Commission will assess the possibility of such effects in the following section.

5.3.4 Effects of the Transaction on new management agreements

(377) Hotel management services are differentiated products. Within a relevant market, products may be differentiated such that some products are closer substitutes than others. The higher the degree of substitutability between the merging firms' products, the more likely it is that the merging firms will raise prices significantly. Conversely, it is less likely that a merger will significantly impede effective competition, in particular through the creation or strengthening of a dominant position, when there is a high degree of substitutability between the products of the merging firms and those supplied by rival producers.

(378) In addition, customers of the merging parties that may have difficulties switching to other suppliers because there are few alternative suppliers or because they face substantial switching costs are particularly vulnerable to price increases. The merger may affect those customers' ability to protect themselves against price increases.

(379) Finally, some proposed mergers would, if allowed to proceed, significantly impede effective competition by leaving the merged firm in a position where it would have the ability and incentive to make the expansion of smaller firms and potential competitors more difficult or otherwise restrict the ability of rival firms to compete.

308 See Article III of Marriott Template Management Agreement (Form CO, Annex 2.2.a.2) and Part V of Exhibit A to Starwood Template Management Agreement (Operating Services Agreement) (Form CO, Annex 2.2.b.2).

309 See responses to Q4 – Questionnaire to hotel owners and managers, question 30.

310 Horizontal Merger Guidelines, paragraph 28.

311 Horizontal Merger Guidelines, paragraph 31.

312 Horizontal Merger Guidelines, paragraph 36.

82

(380) On the basis of those guidelines, the Commission has assessed the ability of Marriott's and Starwood's customers (hotel owners) to switch to alternative management solutions for their hotels.

5.3.4.1 Marriott's views

5.3.4.1.1 Closeness of competition between the Parties' hotel management services and possibilities of switching to other hotel chains for the provision of hotel management services

(381) Marriott has identified a number of other hotel chains than the Parties that provide hotel management services across the EEA, in particular for 4- and 5-star hotels, which the Parties are specialised in (e.g. Hilton, Carlson, Melia, Accor, IHG).

(382) With regard to hotel owners' switching patterns, Marriott provides some examples of hotel owners having changed brands, and lists Marriott's and Starwood's managed hotels having exited certain cities as well as the grounds for termination or non-renewal of the management agreements.

5.3.4.1.2 Barriers to entry or expansion

(383) Marriott takes the view that entry or expansion on the market for the provision of management services is readily accomplished by a broad set of actual and potential competitors. More specifically, Marriott submits that there are no meaningful barriers to the expansion of a hotel management services company to additional geographic areas both within and into the EEA and that the limited barriers to expansion of existing operators (applying to the entire hotel accommodation industry) will not change post-Transaction.

5.3.4.1.3 Competitive pressure exerted by other hotel management solutions (in-house management and white label management companies)

(384) Marriott takes the view that competition for the provision of hotel management services is not limited to hotel chains. Marriott considers that there are two other levels of competition for the provision of management services to hotel owners. It indeed submits that (i) hotel owners can readily choose between managing hotels in-house or outsourcing their management to third-party management companies; (ii) the market place for the provision of outsourced management services is extremely diverse, comprising chains as well as white label management companies.

(385) To illustrate its statement, Marriott gives the examples of hotels that used to be managed by Marriott and were converted in 2014-2016 into franchised hotels managed in-house by their owners or managed by white label management companies.

5.3.4.2 Commission's assessment

5.3.4.2.1 Closeness of competition between the Parties' hotel management services

(386) The market investigation results are inconclusive as to whether Starwood and Marriott are closer substitutes to each other than to their competitors in the supply of hotel management services. Hotel owners and competitors compare the Parties' hotel brands to provide their opinion on their closeness on the market for the provision of hotel management services. As the perception of the closeness of the Parties' brands varies from one brand to another and from one market participant to another, the opinions about the closeness of the Parties' management services are split.

(387) In any case, most respondents to the market investigation questionnaires consider that a number of other hotel chains compete with the Parties on the market for the provision of hotel management services in the EEA (e.g. Hilton, IHG, Accor, Hyatt).

(388) The hotel chains listed as the Parties' main competitors by most respondents present similarities with the Parties in terms of brand positioning, size of the network of managed hotels and level of management fees, which are criteria that most hotel owners take into account for the selection of a hotel management company.

319 Form CO, paragraphs 260 and 261.

320 Form CO, paragraph 153. [Information regarding Parties' hotels that have switched operation type].

321 Form CO, paragraph 160. [Information regarding Parties' hotels that have switched operation type].

322 See responses to Q4 – Questionnaire to hotel owners and managers, questions 21 and 21.1 and Q1a - Questionnaire to competitors, questions 62 and 62.1. Those responses confirm the significant role played by brands in the selection of hotel managers.

323 See response to Q4 – Questionnaire to hotel owners and managers, question 21.1. For example: "Marriott are more of a US brand whereas Starwood are more international. For us, Hyatt or Accor are more of a substitute." vs. "They are all very much on the same level."

324 See responses to Q4 – Questionnaire to hotel owners and managers, question 27 and Q1a - Questionnaire to competitors, question 61. Since the Parties' main competitors identified by respondents provide hotel management services globally, the Commission considers that the opinion by respondents expressed at EEA level also applies at worldwide level.

325 See Section 5.2.3.1.1 of this Decision, which shows that the Parties' brands are close to Accor's, Hilton's, Hyatt's and IHG's brands (in the highest star ratings).

326 At both worldwide and EEA-wide levels, the number of hotels managed by Marriott is the highest, but that of Starwood is comparable to IHG's, Accor's and Hilton's number of managed hotels. In addition, at EEA level, other players (e.g. Melia, TUI, Groupe du Louvre) have a large network of hotels, only slightly smaller than Starwood's network.

(389) In addition, one respondent to the market investigation indicates that neither of the Parties competes particularly aggressively on the market for the provision of hotel management services.

(390) Taking into account all the evidence available to it, the Commission considers that Marriott's and Starwood's hotel management services are close competitors, but are not closer substitutes to each other than other hotel management services delivered by large hotel chains with 4- and 5-star (or upscale, upper upscale and luxury) hotel brands.

5.3.4.2.2 Possibilities of switching to other hotel chains for the provision of hotel management services

(391) While one hotel chain pointed to the substantial costs entailed by a switch of hotel management companies, no hotel owner referred to such a barrier favouring the incumbent provider at the time of renewal of a hotel management service agreement.

(392) In addition, Marriott and Starwood currently operate different systems and brands. Therefore, the Transaction will not have an impact on the potential costs of switching hotel management companies (in particular it will not remove the potential barrier to switching from Marriott to Starwood or conversely), unless and until the two hotel chains adopt the same systems. On the contrary, the Transaction may have a deterrent effect for those multi-property hotel owners that want to multi-source the management services provided to their portfolio of hotels.

(393) Taking into account all the evidence available to it, the Commission considers that hotel owners will not have material difficulties to purchase hotel management services from hotel chains other than the Parties post-Transaction.

5.3.4.2.3 Barriers to entry or expansion

(394) Based on the results of the market investigation, it appears that the element most taken into account by a hotel owner for the selection of a hotel management company is the management company's experience/reputation. Consequently, the entry into the market for the provision of hotel management services requires, as far as hotel chains are concerned, having reached a sufficient degree of maturity in the management of owned hotels and a sufficient degree of hotel brand strength to start acting as an external provider for other hotel owners.

(395) However, once this sufficient degree is reached, a hotel chain does neither face any significant barrier to entry, as shown by the development of hotel chains such as NH Hotels in the provision of hotel management services, nor any significant barrier to expansion, as shown by the examples provided by Marriott and responses to the market investigation questionnaires.

(396) In addition, most of the hotel owners having expressed an opinion during the market investigation consider that the Transaction will not change the level of barriers to entry/expansion in the market for the supply of hotel management services. The opinions are more divided among chain competitors having responded to the market investigation. Contrary to hotel owners, a majority of chain competitors took the view that the Transaction will increase barriers to entry/expansion. However, those competitors did not substantiate their view, other than by referring to the economies of scale brought about by the Transaction and the alleged higher market power of the merged entity.

(397) Finally, the STR database shows pipeline hotels which will be managed on behalf of third-party owners by the Parties, but also by other hotel chains.

(398) Taking into account all the evidence available to it, the Commission considers that the Transaction will not have an impact on the barriers to entry or expansion in the market for the provision of hotel management services and, therefore, that competition on that market will remain dynamic post-Transaction.

5.3.4.2.4 Competitive pressure exerted by in-house management

(399) The market investigation yielded mixed results as to whether the hotel management services provided by a third-party management company can easily be replaced by in-house management by the owner (through its own employees). Most respondents take the view that managing a hotel requires skills, know-how and infrastructure that hotel owners do not necessarily have and may not easily acquire and, more generally, that the feasibility of the in-house management option depends on the hotel owner's profile.

(400) In light of the above, the Commission reaches a more nuanced conclusion than Marriott about the constraint on the pricing and competitive behaviour of the Parties exercised by the in-house hotel management option. It acknowledges that hotel owners could decide to provide management services internally if prices for outsourced hotel management services were to rise above the competitive level, but only to the extent that such internalisation would be compatible with the hotel owners' business model and structure.

5.3.4.2.5 Competitive pressure exerted by white label management companies

(401) In the framework of the market investigation questionnaires, only two hotel owners and one hotel chain identify white label management companies among the Parties' main competitors (e.g. BDL Redefine, Interstate Hotels, Kew Green Management).

(402) However, as indicated in Section 4.2.2, hotel owners regard white label management companies that have reached the required level of expertise as a viable alternative to hotel chains for the management of their hotels (in particular in conjunction with franchising agreements).

337 See responses to Q4 – Questionnaire to hotel owners and managers, questions 7.2 and 7.2.1 and Q1a - Questionnaire to competitors, questions 41.2 and 41.2.1.

338 See responses to Q4 – Questionnaire to hotel owners and managers, question 7.2.1. For example: "The level of expertise required is unique and the hotel is in a different industry to the owner's main business"; "Managing hotels requires specialised hotel management know-how and infrastructure as well as human resources which we don't have. We consider ourselves primarily as property owners.". See also responses to Q1a – Questionnaire to competitors, question 41.2.1. For example: "To manage the hotel, the owner needs to build up the professional management of such hotel, and invest in developing its own distribution channels"; "missing of hotel skills of independent hotel operator; missing of advantages given by franchise company; missing distribution and sales market power".

339 See responses of a hotel owner to Q4 – Questionnaire to hotel owners and managers, question 7.2.1: "It very much depends of the know-how and capabilities and organizational structure of the in-house management" and of a competitor to Q1a – Questionnaire to competitors, question 41.2.1: "Some owners do not have management capabilities and are financial investors only, while other owners have management capabilities. Each owner will have a unique profile."

340 See responses of two hotel owners to Q4 – Questionnaire to hotel owners and managers, question 27 and of a hotel chain to Q1a - Questionnaire to competitors, question 61.

341 See responses to Q4 – Questionnaire to hotel owners and managers, questions 7.2 and 7.2.1 and Q1a - Questionnaire to competitors, questions 41.2 and 41.2.1.

342 See responses to Q4 – Questionnaire to hotel owners and managers, questions 29 and 29.1.

(405) In view of the above, and considering all evidence available to the Commission, the Commission concludes that the Transaction does not raise serious doubts as to its compatibility with the internal market in respect of the market for hotel management services by hotel chains in the EEA or worldwide.

(406) The Commission concludes that, a fortiori, the Transaction would not raise serious doubts as to its compatibility with the internal market in respect of the market for the provision of hotel management services by both hotel chains and white label management companies in the EEA or worldwide.

5.4 Hotel franchising services

(407) As indicated in Sections 4.2.3 and 4.3.3, the Commission will assess the effects of the Transaction on the plausible markets for the provision of hotel franchising services in the EEA and worldwide.

5.4.1 Estimated market shares of the Parties and affected market

(408) The Parties together franchise, across all star categories, approximately […] hotel rooms in the EEA (Marriott: approximately […] rooms; Starwood: approximately […] rooms) and approximately 587 000 rooms worldwide (Marriott: approximately 438 000 rooms; Starwood: approximately 150 000 rooms). Based on the hotels recorded as franchised in the STR database, Marriott estimates that the Parties hold a combined market share of [5-10]% in the EEA (Marriott: [0-5]%; Starwood: [0-5]%) and [10-20]% worldwide (Marriott: [10-20]%; Starwood: [0-5]%). Marriott therefore submits that the market for the provision of hotel franchising services is not affected EEA-wide or worldwide.

(409) The market reconstruction undertaken by the Commission confirms that the EEA-wide market for the provision of hotel franchising services is not affected. Based on the number of rooms franchised by the main hotel chains (e.g. Accor, IHG, Carlson, Choice Hotels, Louvre Hotels), the Parties combined market share would not exceed [5-10]% at EEA-level.

(410) However, the results of the market reconstruction do not enable to exclude that the market for the provision of hotel franchising services would be horizontally affected at worldwide level, where the combined market share of the Parties could reach up to [20-30]% (Marriott: [20-30]%; Starwood: [5-10]%).

(411) Although the market reconstruction exercise is likely to materially overstate the Parties' market share, the Commission considers that the assumption of the Parties with regard to their limited position on the market for the provision of hotel management services cannot be confirmed at this stage.

(412) The Commission will thus further examine the horizontal effects of the Transaction on the worldwide market for the provision of hotel management services.

5.4.2 Effects of the Transaction on ongoing franchise agreements

(413) The Commission considers that the Parties' current customers of hotel franchising services are a priori sheltered against price increases until termination or expiry of their existing franchising agreement, considering (i) the long-term duration of franchise agreements entered into by the Parties (between [...] and [...] years), and (ii) the determination of the financial conditions, in particular the level of franchise fees, for their entire duration without right to adjustment for the franchisor.

(414) The Commission therefore concludes that the pricing effects of the Transaction, if any, may only materialise at the time of award of a new franchise agreement (either for a new hotel or following termination or expiry of an existing franchise agreement). The Commission will assess the possibility of such effects in the following section.

5.4.3 Effects of the Transaction on new franchise agreements

5.4.3.1 Marriott's views

(415) Marriott considers that the Transaction will not lead to competition concerns at worldwide level, since the Parties will continue to face significant competition from other regional and global chains which offer franchising services, and hotel owners can continue to choose from multiple offerings.

5.4.3.2 Commission's assessment

(416) Most respondents to the market investigation consider that Starwood and Marriott are not closer substitutes to each other than to their competitors in hotel franchising in the EEA.

(417) In addition, most respondents to the market investigation questionnaires consider that a number of other hotel chains compete with the Parties on the market for the provision of hotel management services in the EEA, which present similarities with the Parties in terms of brand positioning (e.g. IHG, Hilton, Accor, Carlson). In particular, those competitors franchise 4- and 5-star hotel brands (or luxury, upper upscale and upscale brands), which are the categories or classes in which the Parties are mainly active.

(418) Furthermore, most respondents to the market investigation questionnaires having expressed an opinion consider that, in the market for hotel franchising in the EEA, there will be sufficient competition to prevent the merged entity from raising prices post-Transaction. That general opinion relies in particular on competition brought by other strong hotel franchisors with worldwide high-end hotel brands, as mentioned above. The Commission notes that, in addition to that direct source of competition, the possibility to operate a hotel on an independent basis may also exert a relative competitive pressure.

(419) Finally, although responses to the market investigation questionnaires are divided with regard to the impact of the Transaction on barriers to entry/expansion, nonetheless, given the limited workload requested by an existing hotel brand in order to enter the market in a new location (for instance, the franchisor does not need to provide staff and to oversee the daily operations of a franchised hotel), the Commission considers that the barriers to expand in new geographic areas are relatively low. The market for hotel franchising, which is currently dynamic, is likely to remain so post-Transaction.

5.4.4 Conclusion

(420) In view of the above, considering all evidence available to the Commission, in particular the moderate combined market shares of the Parties and the fact that a majority of market participants indicates that there would be sufficient competitive pressure post-Transaction, the Commission concludes that the Transaction does not raise serious doubts as to its compatibility with the internal market in respect of the market for the provision of hotel franchising services in the EEA and worldwide.

6 CONCLUSION

(421) For the above reasons, the European Commission has decided not to oppose the notified operation and to declare it compatible with the internal market and with the EEA Agreement. This Decision is adopted in application of Article 6(1)(b) of the Merger Regulation and Article 57 of the EEA Agreement.

For the Commission

(Signed) Margrethe VESTAGER Member of the Commission

352 Most hotel owners having expressed an opinion during the market investigation consider that the Transaction will not change the level of barriers to entry/expansion in the market for hotel franchising (see responses to Q4 – Questionnaire to hotel owners and managers, question 22). Conversely, most hotel chains having expressed an opinion stated the opposite (see responses to Q1a – Questionnaire to competitors, question 56). Nevertheless, most of them also consider that the Transaction will not have an impact on the market for hotel franchising (see responses to Q1a – Questionnaire to competitors, questions 65 and 65.1).

353 See for example response of a hotel chain to Q1a – Questionnaire to competitors, question 42.1 on the geographic area on which hotel franchisors compete: "The additional workload for a new franchise contract is low in comparison to the fees generated. therefore many companies compete for those contracts."

354 See for example response of a hotel chain to Q1a – Questionnaire to competitors, question 40.3.1 on whether the market for hotel franchising should be further segmented: "There are enough existing brands (outside of the brands which are the subject matter of this report) to offer sufficient alternative franchise solutions to hotel owners in the EEA."