Description

•      Booking Holdings requires no tangible invested capital, earns an after-tax free cash flow margin in the mid-30% range, has years of high single digit organic growth ahead of it, exhibits counter cyclical tendencies, and has net cash and investments, yet trades at a 6.2% 2019E free cash flow yield (7.5% ex cash and investments)

•         BKNG is cheap on its face because there are several overblown investor concerns including the threat of Google’s dominance of the customer acquisition funnel, increased competition from Airbnb, greater emphasis on loyalty programs at branded chains, the trend of rising marketing costs, and the potential for disintermediation from hospitality software providers

Booking Holdings (BKNG) is a European-focused online travel agency (OTA) that operates in the online travel oligopoly along with Expedia (EXPE) and now Airbnb, too. OTAs are a growing distribution channel for the hotel/travel industry.

This graphic from Infrata gives a complete picture of hotel distribution channels. 

For the uninitiated: TMC stands for Travel Management Company- a B2B company that acts as a travel manager for corporations (like Amex Global Business Travel Agency). GDS stands for Global Distribution System- a network that hotels tap into to share their inventory/rates (examples include Amadeus and Sabre). Leisure Agents are just regular old travel agents. Tour Operators are people that put together entire vacation packages and itineraries which they sell through agents. CRO stands for Central Reservation Office- a call center that a chain would use to handle reservations. CRS stands for Central Reservation System- the piece of the hotel tech stack that manages the availability/reservations for a hotel or chain of hotels. The Booking Platform is the piece of the hotel tech stack that allows for direct booking from the hotel’s website. An Extranet is an OTA’s network (a hotel has a unique login/portal). Wholesalers are people who pay for blocks of rooms up front at a discount and sell them (typically through travel agents) at a marked-up price to earn a spread.

Hotels have high fixed expenses and are incentivized to fill their last rooms to earn incremental profit. OTAs are generally considered a higher cost distribution channel and charge 10-20%+ of the gross booking value, depending on a hotel’s bargaining power. Hotels would prefer to get their customers direct, but in tougher times the OTA value proposition to the hotel increases as the hotel seeks the incremental customer. This leads to an element of counter-cyclicality for the OTAs.

I expect Booking to earn about $5.2B of FCF in 2019 for a forward FCF yield of 6.2%. BKNG has a great balance sheet with $8.7B of debt and $6.3B cash and marketable securities. BKNG has another $8.4B of investments in long term government securities, corporate debt, Ctrip convertible securities, and investments in other Chinese companies (Meituan, Didi). Excluding the cash and long-term investments from BKNG’s market cap, the FCF yield rises to 7.5%.

Booking’s gross travel bookings and revenue can grow at ~10% for the foreseeable future. The travel industry grows at roughly ~2X GDP, or 4-6%. Estimates differ based on the source, but online booking as a percentage of total hotel distribution is currently around 40%. While it’s unclear what market share online booking will ultimately reach, it’s likely to continue taking share for a long time. If online takes 2% market share a year, that equates to another ~5% of gross booking growth per year for a total of 9-11% growth. Yet Booking trades like its either a mature low/no growth company, or it has existential problems. There are several bear arguments, but they appear exaggerated and more than compensated for by the low valuation.

Google Will Eat Booking’s Lunch1

Aggregation Theory says that in an industry with no marginal transaction costs for an additional customer (generally brought about by the internet), the owner of the consumer relationship will come to dominate as supply becomes commoditized. In the case of travel, Google owns a large chunk of the consumer relationship as many travelers start their search for accommodations on Google. This is exacerbated by the fact that many people travel infrequently and are therefore more likely to start their accommodation research where they start researching everything else- on Google. The bear argument for BKNG is that Google’s dominance at the top of the funnel will end with Google either becoming an OTA or relegating BKNG to a commodity-like provider of supply.

However, Google already has the superior economics of the relationship with a high incremental margin revenue stream that grows with the OTAs and requires no additional invested capital. That doesn’t mean BKNG isn’t in a great position too.

Aggregation theory assumes a commoditized supply, but BKNG’s relationships with hotels are not commoditized. Booking primarily serves small, independent hotels and motels, and their average property (excluding the vacation rentals) has only 54 rooms. This compares to branded chains like IHG with an average of 150 rooms per property (100 on the low end at smaller concepts like Holiday Inn Express). It has taken BKNG decades to build these relationships and BKNG is more integrated into these independent properties than is often understood. 350,000 properties now use BKNG’s software solutions, called BookingSuite. BookingSuite provides the property with a website, a booking engine, and data analytics to help the independent proprietor price their supply. Large brands have access to geographic pricing data because of their large footprint, but the independent hotel needs to get it from a 3^rd party. BKNG has the best data because of its scale. With the introduction of the new BookingSuite App Store this month, it’s likely that Booking’s role in the independent hotel’s tech stack will be growing over time. These are not relationships that would be easy for Google to displace.

Booking employs over 24,000 people, two thirds of which are providing customer service in 40 different languages to both the consumer and hotelier. Is that something Google wants to get into?

BKNG now gets 50% of its traffic direct, and that direct traffic is growing faster than the overall company’s growth rate. The more direct traffic to BKNG’s site, the lower its customer acquisition costs, and the more it owns the customer relationship that bears are concerned with.

There are also enduring reasons why the large OTAs acquire most of Google’s travel traffic. Google delivers customers to Booking in three ways: 1) organic results at the bottom of the page 2) paid ads at the top of the page and 3) a metasearch in between (also integrated into Google Maps).

There are technical reasons that the OTAs dominate the unpaid search, but the details are unimportant. Google’s aim in unpaid search is to provide the searcher with the most relevant results, and the large OTAs are the most relevant websites for someone searching for accommodations.

The paid ads will also be dominated by OTAs as they are bid and paid for on a pay-per-click basis. The party that can pay the highest amount wins the traffic. OTAs are some of the most skilled internet advertisers in the world. They are constantly running tests and optimizing ads in an everchanging environment. They are also laser focused on converting someone who clicks an ad into a paying customer. The OTA’s extensive inventory will always be an aid to conversion as a person who clicks an ad for a property that they later decide they don’t want is served up several alternatives to book. These two advantages allow them to bid more than competitors.

If you’re Google and you want to capture more of the customer funnel without messing with organic search or paid results, you might think about inserting a metasearch above the unpaid results. How Google’s metasearch and new dedicated hotel site will affect things is unclear, but there are paid advertisements served first, and the large OTAs will likely dominate those for the same reason that they dominate the paid ads at the top. Booking defines direct traffic as any unpaid traffic, so it is including organic google search results in their estimate of 50% direct traffic. The metasearch between the traditional paid and unpaid results, unfortunately, reduces Booking’s unpaid traffic.

Metasearch companies traditionally took on the role of accommodation supply super-aggregators, collecting and comparing supply available across all OTAs for the consumer. As the OTAs add properties, the value proposition of metasearch declines (I think TripAdvisor’s failed instant booking initiative was recognition of this paradigm). The declining value prop of metasearch will apply to Google’s metasearch too, even if the new dedicated site has disruptive effect on OTAs in the short term. The integration of metasearch into Google Maps is more concerning in the long term. Currently, it’s not very intuitive and it’s questionable whether travelers planning a trip will think to open Google Maps first for their accommodation needs. Will travelers be able to manage their trip from Google Maps like they do from the BKNG app? Will they be able to scan in at events and run their itinerary from one app?

Google does have an element of bargaining power over Booking (as it always has), but it gets paid for that. Both companies need each other.

Airbnb Will Eat Booking’s Lunch2

Airbnb has broken the BKNG-EXPE OTA duopoly. Airbnb announced a year ago that they were going to start signing up hotels/bed and breakfasts, charging them only 3-5% and charging the guest a separate fee too. Bears believe that competition will increase and threaten BKNG/EXPE’s bargaining power over hotels.

Airbnb’s success is proof of how hard it was for a viable competitor to materialize, as it required the invention of an entirely new accommodation category. Airbnb is primarily involved in alternative accommodations. The company took a shot at BKNG in March, citing its more than 6 million alternative accommodation listings. Yet BKNG has caught up, with 5.7 million alternative listings of its own. BKNG recently disclosed that earns over 20% of its revenue from alternative accommodations, and that this segment is growing faster than the overall business. On top of its 5.7 million alternative listings, BKNG has another 23 million hotel listings that Airbnb does not have. All BKNG’s alternative rental listings are instantly bookable (no back and forth with the host) while Airbnb’s instant book page claims that over 3 million listings are instantly bookable (just half of their current listings).

Airbnb has acquired HotelTonight, which many see as Airbnb’s gateway to begin adding hotels to its platform. However, HotelTonight only has 25,000 hotels- less than 6% of BKNG’s hotel/motel/resort properties. Airbnb’s pricing arrangement requires the consumer to pay a greater fee than the hotel, which may work to sign up hotels, but negatively impacts the customer experience. Airbnb’s biggest advantage is its brand name and its higher proportion of unpaid, direct traffic. It seems short sighted to degrade the consumer experience and erode your direct traffic with a fee that favors the hotel, but perhaps that strategy indicates the difficulty of their current position. Airbnb has a long way to go, and for reasons cited earlier, it’s not going to be easy. Others have tried to compete in hotels and have failed. In the meantime, Airbnb is competing against a platform that already has what it is trying to achieve.

Hotel Loyalty Programs Will Eat Booking’s Lunch3

Hotels have been pushing back against OTA’s excessive fees by emphasizing their direct loyalty programs. Customers will recognize the additional benefits of booking direct and the OTA’s market share will shrink.

While it is true that the large international brands have been pushing their loyalty programs, Europe’s accommodation industry is much more fragmented than the US. BKNG represents mostly independent hotels that are much smaller than branded. Large international brands are only 15% of BKNG’s business, and smaller brands (let alone independent hotels) don’t have the scale to operate popular loyalty programs.

Even the largest hotel chains have had a hard time breaking the OTA’s market share despite the additional emphasis on loyalty programs in 2018. OTAs actually grew as a percentage of bookings for InterContinental Hotels (the largest hotel loyalty program in the world) in ’18. Marriott automated their systems in 2018 to reduce reliance on OTAs during peak demand period. They also combined their loyalty programs with Starwood’s in August of ’18 and rebranded them as a single program in early ’19. The result, about which Marriott bragged during their recent investor day, was keeping OTA’s percentage of booking flat at 13% for 2018 (up from 10% in ‘16). These IHG/MAR results also include the direct traffic pick up that they would have gotten from the OTA’s reduction in advertising spend in the metasearch channel during 2018. If loyalty programs and direct booking are threats at all, they affect EXPE to a greater extent because the US market is more consolidated and more competitive than the European market.

Marketing Costs Will Eat Booking’s Lunch4

BKNG’s online/performance advertising costs as a percentage of gross profit have been rising for years. The fear is that at maturity, EXPE and BKNG will bid each other’s profit margins down to unattractive levels.

There are reasonable bullish and bearish arguments for how these companies will look at maturity, but maturity is a long way away. With online booking at only 40% of total bookings, we could easily have another 10-15 years of market share growth, albeit at a lower rate than the OTAs have enjoyed over the prior 10 years. BKNG, EXPE, and Airbnb have intelligent management teams. Will they destroy each other’s margins? We had a hint of what could happen at maturity when BKNG pulled back on performance advertising in Q3 ’17 (mostly in the metasearch channel). As shareholders would hope, EXPE followed BKNG’s lead a few quarters later. BKNG’s performance marketing as a percentage of revenue has been declining ever since. This won’t continue forever, but it does give some comfort that the OTAs will act rationally as they approach maturity.

Hospitality Tech Consolidation Will Eat Booking’s Lunch5

Jonas Software is rolling up hospitality tech providers (Leonardo and BookAssist). Shiji has acquired StayNTouch, Concept Software Systems, is backing Kalibri Labs, and has recently opened their distribution platform to international markets. Cloudbeds is one of the fastest growing travel companies of 2018. Amadeus acquired TravelClick to augment its strategy of offering hospitality providers a full suite of software solutions including a CRS, property management system, sales and catering solutions, and a payments platform. This consolidation could result in a single dominant software provider that connects hotels into its distribution network and disintermediates the OTAs.

It seems too early to worry too much about this particular risk. The offering from Amadeus is aimed at hotels much larger than the average hotel on BKNG’s platform. If something like Cloudbeds really takes off, BKNG could potentially acquire them to help cement their position at the small hotel. BKNG’s RateIntelligence pricing tool is going to be very difficult to compete with because Booking has extensive scale and therefore better data. Booking also has the advantage of being able to offer BookingSuite for free, bundling it with its traditional OTA services, while a small hotel would have to pay for a competitor’s software suite in addition to BKNG’s traditional commissions. It’s fair to say that BKNG is keeping an eye on competition in software and the introduction of the BookingSuite App Store is an attempt to compete.

Is Anyone Going to Eat Booking's Lunch?

Maybe, but it seems unlikely, and the stock is very cheap for a company with such a high-quality financial profile.

Appendix A: Memes

1) 

2)

3)

4)

5)

Catalyst

Time