Booking Holdings Inc. (BKNG) Earnings Call Transcript & Summary

All right. Good morning. Welcome back to the next session. My name is Lloyd Walmsley. I run the Internet research team here at UBS and excited to have Booking Holdings for our next session. We have Glenn Fogel, CEO; David Goulden, CFO. Guys, thank you so much for being here. It's great to have you.

Thanks, Lloyd.

Great to be here. Thank you.

Yes. Well, look, we're excited to jump in, and we're going to start at a high level. There's this view out there, some executives in the space kind of trumpeting more than others that work from home kind of permanently expands the addressable market of travel and may change the demand mix between hotel and non-hotel. Glenn, where do you come out on that? What do you think travel looks like post-pandemic?

Yes. I think that definitely, there's an expansion of the TAM. I don't think anybody knows yet how big, how much, but you end up with these flexible work environments and that's just going to create more travel opportunities. So people are going to take mixed trips. They're going to do work away from home. They got to go somewhere to be away from home, they got to actually travel. And then you got to know, what kind of place you're going to stay in. And there's more bleisure, people who are going to say, I'm going to go away for Thursday, Friday, Saturday, Sunday. So I'll work some of the Thursday, Friday, but I will also have some fun too. So there'll be more of that kind of thing. So that's definitely happening. And then you go on to like people saying, "I'm going to work away -- really away, I am going to be in another country, like, why do I have to work in Amsterdam. I mean it's dark and cold in the winter, I'm going to go down to Majorca, and I'm going to go work from there. And that's going to happen much more than it ever would have happened in the past. That's more travel there, too. So a lot of positive dynamics. That's really, really good. And people are going to do -- it's not just home. I want to make sure everybody understands that the hotel people, they are not dumb, they understand this trend and they're trying to come up with ways that they can make sure that their product fits well for people who want to do that kind of travel. So that's really good for us and the travel business, I like it. Now I don't know how much. And by the way, there's another thing, too. Once you're working away from home, then occasionally, you do have to go to the office. And if you're far away, you got to travel and stay. We're not sure how much, how often, nobody knows yet how it's going to be. But you go back once a quarter or once a month, there's more travel there too, I don't know how big it's going to be. But this whole thing, it's very uncertain, as we all know, everything seems to be a little bit uncertain right now. But I do believe it's going to be a good thing for us. We want to be part of it. We want to make sure that we are helping our customers be able to do this easily, get great value for it, and we want to work with our suppliers, make sure they're pushing out there what they are offering, making sure that people who want to travel see those offerings. So all-in-all nobody likes a pandemic, let's be serious. That's a terrible, terrible thing. But learning that, hey, we don't have to work 5 days a week in an office. It makes life a lot better, I think, for a lot of people.

I want to talk on near term with -- just get it out of the way on Omicron, the latest strain and some travel restrictions popping back up. Is there anything you guys are seeing in bookings or cancellations with the latest news?

Look, it's very early, too early to really talk about this. Let's wait -- let's see what the scientists come up, but see the questions that we all have, how effective are the vaccines against this variant and how transmissible is it? Do you get sicker? Or do you not get as sick? Lots of questions. Everybody wants to know the answer to, and there are no answers yet. So let's find out before we all start thinking what's this really going to mean. I will say though, as we all know, there have been some actions by governments already in terms of this and making some things people. Everything from countries say, we don't want any travelers right now to, hey, we want to have quarantines to well, we want more testing and all different things. And any time a government goes up with any sort of a change in terms of adding friction to travel is a negative for travel. It's that simple. And when they go the other way, it opens up and people travel more, and we love it. And we see it. I mean I know -- I can know most instantly when a government has let down a restriction because we'll see the bookings go up. So this is no different than throughout this pandemic, we've noticed this ups and downs, places raising barriers, people lowering barriers. We'll see. The one thing we do know, though, this is not -- is absolutely not the spring of 2020. Absolutely not. Look, we got vaccines. We have a very rapid ability to come out with new formulas for vaccines. We've got all these antivirals coming out that look very promising. Lots of good things happening. Long term has not changed in the least, long-term travel comes back. We live like we used to in terms of travel, that's actually going to happen. It's just a question of timing and nobody knows the answer yet.

So now that we've had a couple of these strains and these patterns, like do you think people are getting kind of to the point where they're learning to live with this and travel through it? Again, maybe it's too early, but is there any signs as we've gone through this that give you hope that the next 3 strains, it may just -- impact may be less and less?

Well, I think over time, people get used to anything, and you see that in all types of events that happen that cause change and people get used to it over time and start operating differently and accepting risk that they, at the very beginning, because it was so unknown, that they were more scared of it, then they become less scared of. And we see that all throughout. And just a perfect example of that will be yesterday, and I don't even know the number, what 1,600 people died in the U.S. maybe from COVID, I'm not really sure what the number is. The reason I'm not sure is because I didn't see it blaring on the front page in giant type on the newspaper because we've gotten used to, okay, people are going to die and people have gotten used to that. That's a perfect example over time, people are getting used to it. I took a train down to Washington last week, and everybody is supposed to be wearing a mask, there were bunch of people who weren't wearing mask, they didn't care. And you're seeing all that people getting used to it and people just saying, "I'm willing to accept the risk. So I think it's not a linear recovery, and we said from the very beginning, it would not be, they'll be ups and downs, ups and downs, time when you're more fearful, times when you're less fearful, but the trend is absolutely going the right way. People will get used to this.

So stepping back from this specific strain, given kind of the strength of your balance sheet, your long-term focus, like do you feel like you're picking up market share in -- whether it's this strain or just throughout the pandemic in general, given how strongly you're positioned going into this?

Yes. We talked about in our call -- third quarter, how we are pleased with how we are doing. We like what we're doing, particularly in the U.S., we see where we're picking. We think we're picking up share. And we like the way we're doing it. Working very hard with our partners to come up with the right value from them that we can then put out to the travelers and making sure the travelers know about it and making sure they market it appropriately. So all good things happening there, and that's the way we're actually going to continue to go forward. There are no silver bullets here. It's not like some magic thing today to do, over time our magic is our connected trip, that's going to take time. We're continuing to invest in that. I'm not pulling back at all on that. But it's definitely going to be something that's a lot for now just blocking and tackling, and I am very pleased how we have performed over what is approaching 2 years of pandemic. I think we've done an excellent job of managing the business, managing all elements of it and making sure that we are providing the services and the help that is needed for both sides of the marketplace. And Dave, I don't know if you want to add anything to that?

No, that's a great answer.

You guys have announced 2 acquisitions recently. I wanted to just get your thoughts on how -- maybe starting with the Etraveli, how this Flights acquisition fits into your air strategy?

Well, so we've been working with Etraveli for a number of years. So it's not like we just had them yesterday. We've been working with them. And all the great things we've been doing with Booking.com on flight, that's powered by Etraveli. Booking.com doesn't have the relationships with the airlines, doesn't have all the infrastructure, all the connectivity, all the incredibly good pricing stuff that Etraveli is putting out. There are a tremendous amount of technology -- real intellectual property, there is relationship, all sort of things they've done. And that is what has been enabling us on the front end to be able to market and put it out to people and start to build that Booking.com flight, which is so important for the connected trip that we've talked about. And seeing those numbers intact. So it's natural having seen how well they are performing and such and being able to say -- there are certain things we can't do right now because we -- 2 separate companies and the restrictions in terms of how we're doing things, it's contractually that we agreed to do because Etraveli as an independent company had to make sure that they were protecting their own future, et cetera. Putting them together, is really, I think, going to enable us to accelerate what we're doing and sees us the incredible bother and trouble of building up a whole back end and all the things I just talked about ourselves in to build it, we can get it right away. And of course, obviously, we'll be able to put all of our attachments like hotels and cars and all the things that we have into anybody who buys a flight through Etraveli. So this is a real win-win, I think, for all of us, obviously, not done yet. We have to get through the regulatory environment, all that takes some time. I'm really, really positive on this. So I think it's going to help us as we continue marching along our connected trip vision.

And what about Getaroom? Can you just talk a little bit about what they bring to the strategy and how you're integrating them or planning to integrate them?

Sure. So we have strategic partnership businesses in all of our companies. The price line, it was PPN, it is called a Priceline Partner Network. And Getaroom is going to fit in with that. And Getaroom does a bunch of things that we don't do. For example, they sell stuff on the phone. That's something we don't do. And they have a very sophisticated way to bring in supply from lots of different inventory sources, including, by the way, Priceline.com has been supplying inventory into that for some time. Again, a group that we've known for a very long time, good partners that we work with, and we like a lot. Again, bringing them together into the Priceline company will enable us to do things we couldn't do before, enable us to perform better in terms of services to the affiliates and help provide a better service to them. And also really important is Getaroom has some contracts, has some affiliates that we didn't have and weren't going to have and getting them to be part of that, that's a great addition to. So overall, a plus-plus for both Priceline and Getaroom and the affiliates who are getting all the inventory, et cetera. So this is a really good thing for us.

Connected trip, when you kind of think about this strategically, how much of this is designed to increase revenue per transaction versus really just an effort to drive people to mobile app and book direct or maybe a little bit of everything.

Look, every single one of these things is important. And obviously, we get flight, so we get more new customers who are coming down the flight path never would have come to us just for a hotel, but they come down the flight, they get the flight and we give them and they're getting the hotel too, and that's really, really good. And then they get in a different way and they get a car thing. And they get it in a sophisticated way that provides them with the absolute best thing. And they find out when they're doing it on the mobile app, then they find all the attractions while they're there. Because they're carrying their phone with them on this trip, and we're giving all sorts of value there. And of course, it all gets tied together. You got to have a payment platform. It's all tied together in a payment platform. We can do all sorts of merchandising that we couldn't do before. So all these things are going to do all the things you talked about. It's going to absolutely bring in higher revenue. It's going to bring in more customers that we wouldn't get before. It's going to increase loyalty. It's going to get them going through the app, which is a really, really great way to do everything in travel. Win-win-win all across the board, very pleased about it. It's going to take some time, though, but I'm very happy seeing the progress we have made so far.

Are there any north stars we should think about in terms of the long-term attach rate of things like air, car or experiences?

Yes. So of course, what I would like my fantasy -- and it's a fantasy of course, but fantasy is that everybody buys everything always from us and no one else, and that's what we want for everything, of course, that is not going to happen. There are some unique things that people are going to want specific other people to do so. But really -- I'm not going to give out any numbers or not, but I really believe that we wanted to do much, much better than the average industry benchmarks are right now that people see [ and not going to tell me ], you can go look them up for yourself. But I absolutely believe that providing a much better experience will help develop that loyalty, people coming back and then always buy all as many products as we offer them because it is a better experience, provides more value, creates a better customer, issue -- if there is any sort of problems that we fix it much better. Look, and I say this all the time, and everybody on this call knows this. Right now traveling is not nearly as easy, as good as it should be. We've all had terrible frustrations, problems in all different ways I can list it for an hour all the different problems I've had over the 20 years of traveling, well, a travel company, god knows what it is for people who are in the travel companies, and we've got to improve that. And by improving it, by doing and making it much better that will build all the things that anybody who's trying to sell something is, which more value, better experience and being sure that if something goes wrong, you're there to fix it.

So looking at the mobile app, we've seen in third-party data, you all talked about a healthy increase in app traffic. And so aside from kind of just the consumer shift to mobile, what are some of the things you all have been doing to drive more of that?

Yes, Lloyd, let me jump in and take that one. So a couple of key things. First of all, the app experience has to be great, right? It has to be something that customers enjoy in doing. We offer great value to customers via the app, and it's a good experience and a simple tip to use and easy to come back to because the difference between the app and the traditional booking interface is really very important because a traditional booking experience is you make a booking, you save it away somewhere and you're basically done, you may be print it out and take it with you. But with the app, we really want to interact with our customers and join the trip as well, which is why it's a primary interface for the connected trip in the future. So we've been doing a few things. Most of our advertising and promotion has been tied to people coming on to site using the app as we're pretty sure we launched the back to travel campaign in the U.S. this spring. We actually rolled the same campaign out throughout Europe as travel recovered in the early summer. And the way to really get that promotion was to download again or re-download and book via the app. So it's certainly a key area of focus for us. You mentioned the fact that we're getting a number of downloads. That's good. We also mentioned the fact we crossed the 100 million monthly active user mark last quarter. That's also important. And finally, just in terms of the different ways you interact with us directly, you can interact with directly via the app, via mobile web and via the desktop, they are all 3 ways to come on to site directly. Not too surprisingly, the app has the best return to direct characteristics. So it's the stickiest of those ways of dealing with. So we're encouraged by the progress we've been making here.

So when you look at the success you've had in mobile app and increasing share of direct, does that change how you think about spending in performance channels that greater lifetime value? Or are you still focused on kind of transactional unit economics and performance spend.

Yes. We are principally focused on the unit economics, as you mentioned, our ROI is kind of based upon the transaction. However, we do use lifetime value as a secondary metric, and that can help us to determine how heavily we lean into one channel versus the other based upon the channel's propensity to give us customers that we can continue to keep on the platform as future or customers of ours.

You guys talk about it a little bit, but not, I think, as much as you could, but the Genius program, I think it's over 70 million members. It's been a key driver of loyalty. How has the program evolved over the years? How do you see it evolving over the next couple of years as you add more product into the mix?

Yes. Sure. So let me, first of all, talk a little bit about Genius because it really is an important program. So at Booking.com, of course, it's a great example where we have listed hundreds of thousands of our property partners participating, offering value to our customers in the form of lower rates, complimentary breakfast, room upgraded -- room upgrades. We have things like discounted airport taxis. And we've recently also extended Genius customer rates on some of our rental car partnerships as well. So the benefit of being a Genius member continues to grow. The 70 million stat you gave was actually back to 2019, and that was just the Genius customers who actually made one booking in the prior 12 months. So the total Genius lifetime base was much larger than that. Since 2019, we've also expanded the program a couple of ways. So first of all, now you get to Genius Level 1, if you're a logged in active customer with an account. So we've expanded the size of the program quite substantially. I haven't given our number out, but it's a fair amount, bigger now, because anybody who's an active long gone customer now get Genius 1 benefits. And you'll also start us to see in the not-too-distant future, are rolling out a new level 3 of our Genius program with either further benefits with even more frequent customers. This actually ties back a little bit to what I mentioned about direct and app because frequency is also another key attribute here. Not surprisingly, the more frequent a customer operates with or works with us, the more often they're going to come back directly as well. So this is another way to kind of encourage customers to continue to use the platform more and more. And of course, as we expand out through the connected trip, all these things can lead to a solution where we're looking to get more of our customers buying more of their travel from us, using the app and use us more frequently, that in turn, should generate more loyalty and more direct business.

So looking at some of the fintech stuff you're working on, you said last quarter, about 1/3 of Booking.com's bookings are now running on your payments platforms. What are the key levers for getting that to over half of total bookings? Are there chunky step-ups of adding different chains? Like what are some of the things that will drive that penetration rate higher.

Yes. So we've obviously moved up rapidly. We were only 15% in 2019. So from there to 1/3 is a pretty big step. In terms of steps to increase, 2 sides. I mean, one is obviously enabling our partners. More of our partners are enabled than the 33% of bookings you see that actually transact. So I'd say it's more -- driving adoption is a bigger factor than enablement. There are still some partners we're working on enabling, but the bigger driver would be to drive adoption up. And that's up to both our partners and customers. We make it relatively easy for our partners to opt into our payment platform once they have enabled themselves on the platform, decide which rates they want to offer on a payments platform with us taking the payments as opposed to them taking payments. And of course, on the customer side, continue to make it attractive for customers. We do find there's some positive conversion benefits and even offering different alternative payment methods, even if people don't actually take them. And then, of course, the whole payment mechanism and process is designed to take a lot of friction out of the booking process because the agency model is inherently very simple and very attractive. But actually, confusion by both the hotel and the booker in terms of how the payment transaction happens, is 1 of our biggest sources of customer service challenges, and that's 1 of the things that's also addressed via payments when we do it ourselves.

Okay. And then as we think about the revenue model, is it -- today, is it fair to say that most of the revenue generation on the payment side is rebates from virtual credit cards? Or are you starting to ask suppliers for higher take rates in merchant since you do absorb the cost of the credit card and then some customer support? And like how does that evolve over time?

Yes, at the moment, it really is based upon virtual credit cards, bank transfer fees, breakage, the more traditional streams in having a payments revenue flow. But you are right, we are taking -- we do take on credit card risk, full risk, we take on the cost of card-not-present authentication, et cetera, which is an important service that we provide. At the moment, our focus is really driving penetration. That's why we're growing those revenue streams in approximately operating breakeven right now. But there are potential for us to -- I wouldn't say charge more on the take rate, but certainly charge more for value-added services. We can build on top of the platform, things like foreign exchange services for both the booker and for the hotel partner, pay in your own currency, Buy Now, Pay Later, things we can do around our digital wallets, et cetera, there are all things we can kind of build on top of that base platform and start to mine size them over time.

Yes. So one of the things, I guess, that seems really compelling about it is just this notion that interchange fees for hotel are so high for cross-border, especially for multicurrency. And historically, at least if we go back to '19, I'd imagine these are really high percentage of your transactions for these cross-border, cross currency, where you can save the hotel a lot of money and over time, maybe make some money as you do your own payments on these transactions. Like how meaningful is that in a normalized travel environment that kind of current -- what percent of your transactions, how high is that, that are multicurrency?

Well, if you remember, we said over 50% pre-COVID, about 55% were related to cross-border travel. Now some of that is obviously within Europe, right? So you have a common currency there. But I think, unusually compared to most payment platforms, we have a very high mix compared to most of cross-border travel. And as you say, Lloyd, that's where there's opportunity to reduce fees, reduce friction for everybody taking benefit of the size and scale of our payments platform. Obviously, the amount of that we're processing through our platform dwarfs that of almost all of our supply partners, maybe some of the very biggest changes you've got decent economics as well, but both of our partners don't have that kind of scale. So that is obviously an opportunity for us and makes the payment asset we're building here, I think, quite attractive.

And then as we look out a couple of years as you scale this payments business, I think you've talked about margins shaking out somewhere between flight and accommodations. And so I guess, aside from driving scale, what are the other key buckets for getting this margins up inside payments?

Yes. It's things that we just talked about. It's leveraging the additional revenue streams that I discussed. At the moment, our primary focus is as an enablement for the core business. I mentioned all the points of friction we can now resolve by having a good payments platform. And then the margin opportunity comes from once we turn on some of those things that we haven't got turned on today in terms of the monetization capability on top of that. And also, there are enough movements up and down the payment's income statement itself just in the basic transaction side to start to move a little bit towards margin as and when we decide that is the right time. But most of that is going to come from some of these newer things that I just talked about, some of the value-added services we can build on top of the base platform itself.

Okay. Anything you can share with us on when you guys would look to roll out some of this value-added stuff that will be helpful there.

We've already started, I'd say, testing some of those products, not scale. We'll be building the capabilities. So we can start rolling them out. I'd say that we're not at the stage where we want to do a full-scale rollout yet, we'd be building some of the technology and testing the conversion factors. So we've been doing the groundwork. I say we're still in the building mode or the kind of scaling mode of payments certainly for the next 12-ish plus months, but maybe towards the end of that time frame, we could start monetizing some of those things, but not just to the short term.

Got it. And then you touched on this earlier, David, but you've always benefited from having really high conversion rates from your, I guess, paid marketing and organic. And I think part of that has just been the agency model, having liberal cancellation policies. But you mentioned that just adding new payments is actually positive for conversion. So like I guess if we look at payments and then connected trip more broadly, are you seeing any deterioration in conversion as you increase the complexity of the product? Or is it pretty stable or improving?

I think we can do things with payments and with the connected trip combined to increase conversion. We can do credits and discounts with the payments platform we can offer if it's just an agency product, that's up to the hotel property, just to take advantage of the tools and things that we've given them for them to market on our platform. We can participate more directly. I also mentioned that just by having familiar payments alternatives available on the platform. Even if people don't book through payments, we've noticed first of all, but a notional increase in some countries in conversion because you're showing them something they're more familiar with. Also as we introduce new products, one of the things I didn't mention before, buy now pay later is something also we're doing a little bit of [ thing ] right now. That's something that we think can also be attractive for customers down the road as an additional capability on top of Payments platform. So no, I think that it gives us different ways to participate, certainly shouldn't be a reduction of -- on conversion, and we're not trying to make it more complicated. We're trying to actually make it easier for the customer to kind of want to deal with and continue on the platform. And then going back to what Glenn said, Payments really underpins the whole connected trip. It's very difficult to think of a great payments or great connected trip experience if you didn't have a connected payment experience. Because one of the things that we can do, whether it's pay now or Buy now, Pay Later, however, you're handling it, we will handle the distribution of funds to the different partners. And you don't have to worry about having to write multiple checks or multiple credit cards or however you want to pay, we take care of that for you via the payments platform. So that's another place that the payments comes together with the connected trip and also ties back to what we're doing with the app as well.

So shifting to the marketing. How do you see kind of the industry right now in terms of where on the spectrum things are between kind of growth and profitability in marketing channels.

Well, that's a broad question. I certainly wouldn't want to speak for others in the industry, but I would just say that the marketing channels, the ones you're talking about, continue to be competitive, and we expect it to be competitive going forward. In terms of what we saw recently in the first half of this year, as things were still recovering, we saw higher ROI in our paid channels. That was a good thing. And as we expected in the third quarter, we saw slightly lower ROI than we did in 2019. We signaled that at the start of the quarter, and we invested into capturing more demand during the peak season. We think that these marketing channels represent a good way to pick up incremental demand in the marketplace, and we'll stay disciplined and may maintain a strong ROI approach to our spending, something I think we do quite well. So the final point I'd say is our strategy relative to those channels hasn't changed. We continue to seek high-quality traffic at the right price. And to us, high quality means high converting traffic. You talked to conversion with a lower probability cancellation. Of course, we do have a high percentage of our bookings are cancelable, but we model what we expect and a high probability of it is coming back again on a direct basis.

And I guess you guys have started to talk a bit more about spending in social. So is that a function of a big ROI unlock or more strategic shift in how you think about ROI or the funnel.

Glenn, why don't I hand that one back to you.

Had to go off the mute there, so let me speak. Yes. So we certainly talked about this a bit that we want to be expanding to other channels where we know eyeballs are. And people are looking. Now the amount we've been spending is relatively low compared to the overall marketing spend. And we are continuing to develop and we talked about, and look, we're looking at the ROIs closely. We want to make sure as always we want to spend correctly. We don't want to wait. I never want to waste money, but you have to put money to work and experiment and optimize and try and improve them. And we are pleased that we are seeing ways to improve the measurability, which is the first thing. Do we believe the data, do we believe it's right. And we're doing it in a privacy safe way because we know we're all talking about that, too. Make sure we're doing this stuff the right way and the smart way in a way that will actually last and be able to be what customers are cooperating with. We're lucky. We always -- we mostly use our first-party data, and that is obviously a lot easier, you don't have to worry about some of the changes that have happened recently. It's not really impacting us at all. And we're using that to improve incrementality. So we're going to continue to work on this. I would love to be able to spend a huge amount of money in the social channels with the appropriate incremental ROI. And we'll continue to work on it. There's no reason we should in the long run. Look, people are there, people are -- that's where they're spending their time, we should be able to provide a marketing message to them that gets them to come back and buy and that's what we're working on.

You just mentioned this, Glenn, but just to ask it more directly, the -- any targeting or attribution changes from how Apple has changed, it used to be IDFA and ATT, and kind of how you think about these newer social channels in particular?

Way back in April, we talked a little bit about some of the changes, what was going on with Apple. And we decided that we're not going to show the app tracking opt-in prompt. We said, no, we're not going to do that. And we said, look, what some other people use third-party cookies and things like that, that's not really where we're at. Our primary focus, first-party data, leveraging that data in our marketing to make sure that we are able to come back or we were able to operate our marketing machine in the right way. I'm very pleased with the way we've been doing it. I don't think we are impacted very much at all by some of the changes that Apple has done that has disrupted some of the people in the eco -- of the advertising ecosystem. Some other people may be having some issues, we're really not. And what we do believe -- I just want to make this very clear. We do believe in doing what is right in terms of privacy. And we believe you can provide a great targeted marketing message in the right way without having to do certain things that people weren't comfortable with.

So what is the latest in terms of how you think the EU kind of digital market stuff is going to potentially change the landscape with big tech and at booking in particular.

Well, I think the first thing is nothing has been passed yet, and it won't be for some time. There are different stages, things are going through. And eventually, you've got to have the EU governments themselves had to do some action. So still a lot of steps between today and what the actual legislation is going to be at the end. So until then, we don't really know a lot. We can say, though, we can say that if we were to be named, for example, a gatekeeper. Well, we're still not sure exactly what that would mean and there's also going to be a process that we can go through and say, "Hey, we're not a gatekeeper even if we are initially a designated gatekeeper on your certain types of quantitative metrics that are being used to say that's a gatekeeper. So we will always have that process afterwards to then go through and say we're not. And what this can do for our business, nobody knows until the laws come out. But it could be positive, it could be negative. We don't know. I will say this, 2 things. One, it's a little bit disappointing for us to even be considered in the possibility or probability of being a gatekeeper in the initial designation, given the incredibly high competition in the travel industry, particularly in the combination area, particularly in Europe, where we are a relatively low share of this, and everybody knows. You want to buy a hotel or a home in Europe, there are lots of different ways to do that. So the idea that we are gatekeepers, to me, we're not. I strongly disagree with that. Then the next thing is, how it's going to -- actually it could impact some other people in the industry much more than us, which could actually be benefit to us. We'll see how it plays out, and we'll just have to wait until things like Google, what does it mean for them and how people can use Google or not. Who knows, we'll find out.

Shifting to alternative accommodations, is rent by owner supply, like how big of a strategic focus is that? And what are you guys doing to try to get more of that sort of supply onto the platform?

So look, we believe in the U.S. there are, look, a significant percentage of the single property owner revenue opportunity we can get that through professional managers, which is where we're really focusing on. Our first thing right now is to get that using professional managers, it's better -- it's a better way to do it than having to try and gather a whole bunch of single property owners 1 by 1 individually. We're working on that because we know to build out that home product for us in the U.S., we need more supply, and we've talked about that. We're going to continue to work with that, and we're going to do it by focusing on the professionally managed area. We've only been doing it for a few years in the U.S.; Europe, we are much further along. We've got a great alternative accommodation business in Europe. We've been doing it about 15 years, it's really good. Here in the U.S., we have some things we got to work on to make it better and be more competitive, and that's where we're spending time, energy and effort. And I absolutely am pleased with the progress we're making, and I think we will continue to do so.

If you look at supply more broadly, we haven't seen the same growth like we used to. Where are we, do you think, in terms of just getting hotel and alternative supply? Is there a lot left out there to get over the next couple of years? How do you guys think about that?

Yes. So we have 2.4 million properties. That's about 28 million approximately listings. That's an awful lot. Now -- but it is not every accommodation. There are others globally that we don't have yet. So there are some things we can still add on. But I agree, it's not like you're going to triple that, you aren't going to double that. But -- so we think -- it's not like we need more supply. Right now, you may know some people are traveling because of this COVID thing. So supply, supply overall is not -- there are areas, like I mentioned, the single property stuff in the U.S., yes, we want to build that a lot. Absolutely. But the big picture, no. I would say that we're okay, generally speaking.

Okay. And then looking at online penetration, do you think this is something that the e-commerce pull forward can actually help travel. I mean, historically, travel has been among the more penetrated areas online. But do you think that just there are some incremental people pulled in from the pandemic who will be new to travel. How do you guys think about that?

I think we do, Lloyd. I think that if you look at the third-party data, online penetration for combination bookings globally, prior to the pandemic was around 50%. And that have been increasing roughly about 1% a year being kind of slow steady increase. In terms of the impact the pandemic has had on the industry, it's been much more impactful on the off-line sector than it has on the online sector. Now a number of the smaller off-line players have disappeared, a number of bigger ones that you have cut back on our brick-and-mortar capabilities. And the recent estimates would indicate that, that percentage share from 2019 may have actually increased by quite a few points -- a few points from 2019 to 2021 compared to the 1 point a year we've been seeing before. We also have seen good survey work shows that of the people who come to travel new during the pandemic because people have gone to almost everything online during the pandemic but travel maybe one of the stickiest things that they're buying in terms of accelerated online shift that will occur afterwards. So we don't see that percentage going down again. Maybe it goes back to its kind of small increase. But I do think there's been a step increase in online that we've benefited from -- all the online players that benefited from during the pandemic, but it looks like it is here to stay.

Thinking about inflation, I wanted to ask a few questions. First, where do you think ADRs now are where you -- in, I guess, across both the hotel and non-hotel, where are they now versus where you thought they would be at this point in the recovery? And do you think that those ADRs kind of within the same channel mix are likely to continue to go up into this kind of peak summer 2022 season?

Yes, this has been a very interesting phenomenon to see happening because what we're seeing happening right now is a little countercyclical because historically, during a downturn, ADRs have been a lagging indicator compared to occupancy, but even though occupancies are still below on average where they were in 2019, ADRs are increased. So what's happening? So we're seeing a couple of things. So if you go back to what we said about Q3, our constant currency ADRs up about 10%. About half of that was due to mix, less Asian, more North America. But the other half was just due to increase in rates, principally in Europe and North America. And we're seeing 2 things contributing to that. One is that you can't look at the averages for everything. There's certainly strong leisure demand in certain locations. And that has put a lot of pricing pressure on historically leisure-oriented destination. So for example, if you want to go to travel and find a hotel in Miami for Christmas this year, good luck, right? You'll be paying 150%, maybe 200% what you paid for in 2019. That's just supply and demand. But there is -- I think the other fact is Lloyd, it comes back to your question is we do see the cost is one of the things that's putting pressure up. Cost is impacting profit of clients in a couple of different ways. One is they are getting general inflation costs and they're getting labor cost increases, but also, in some cases, the challenge with labor availability. So the occupancy actually may be less, may be higher than it used to be. If they've only got half a hotel open because they can't get labor to kind of keep the other half open, they can charge a lot for what they actually have in terms of inventory available to our customers. We said on the last call, as we go through into Q4, we expect a little bit less of an increase. Regional mix will normalize itself out but also occupancy costs will go down after the peak season in the summer. But some of the underlying things that I talked about, particularly on the cost side, certainly would appear to be here to stay.

And if we think about that cost side and your own P&L, maybe you can walk us through and remind us a little bit, you took a couple of points out of variable cost using the '19 base. I think some of those should start to come back and then you've got some inflation. What are some of the dynamics on maybe personnel costs or through the P&L that we should think about exposed to just this trend of inflation?

Yes. So I think -- yes, we are seeing wage inflation. And obviously, some of the technology-related roles, that is an industry thing that I think has been well talked about. Perhaps the best way to look at this is just to kind of -- if you look at where we were in Q3, the quarter just ended, versus Q3 in 2019, that was obviously before we've done many of our reductions. We had -- our salary expenses were actually down just taking all -- just going back to base salary. Our salary expenses were down compared to where they were 2 years ago, but they're down less data high count. So that's just the impact of a couple of years of merit increase because obviously comparing 2019 versus 2021. So we've had merit increases, normal market pressure, cost of living going through in early '20, early '21, but also reflects the fact that we've also had to respond to some of these underlying cost of labor increases from a competitive point of view in terms of how we think about some of our key labor pools. And I think that will be a factor we'll have to look at again going into '22.

So last one to wrap up would just be capital allocation. Anything changed? You started to give some color on the buyback again. Anything changed with the latest strain? And then if we think out back to a normalized environment, what do you -- you generate tons of free cash, how do you guys think about M&A versus share repurchase over the medium term?

Let's go to the medium-term answer first. We do believe that return of excess capital, capital over and above what we need to do to support the business through organic and inorganic acquisitions is a good part of our overall value creation strategy for our shareholders. And we believe that, that should continue to be part of our equation for increasing shareholder return. In the short term, we said that we plan to start returning capital in early 2022 assuming this travel recovery continues. A new variant doesn't necessarily change our thinking. We said it would be our nonlinear return to loyalty. But as Glenn said, over the next couple of weeks, hopefully we'll start to see some more indicators to where the variant we're looking at right now is just another strain or something different.

All right. Well, Glenn, David, thank you guys so much for being here. It's great to have you. And hopefully, you guys come in the real life next year, fingers crossed.

I don't think we need to cross fingers. I'm highly confident that we'll be doing it next time.

Look forward to being there.

All right. Well, thank you, guys.