萬豪國際 (MAR) 2019 Q4 法說會逐字稿

Ladies and gentlemen, thank you for standing by.

And welcome to the Marriott International's Fourth Quarter 2019 Earnings Conference Call.

(Operator Instructions) Please be advised that today's conference is being recorded.

(Operator Instructions) I would now like to hand the conference over to your speaker today, Arne Sorenson.

Thank you.

And please go ahead.

Welcome to our fourth quarter 2019 earnings conference call.

Joining me today ar3e Leeny Oberg, Executive Vice President and Chief Financial Officer; Jackie Burka McConagha, our senior -- new Senior Vice President, Investor Relations; and Betsy Dahm, Vice President, Investor Relations.

Let me remind everyone that many of our comments today are not historical facts and are considered forward-looking statements under federal securities laws.

These statements are subject to numerous risks and uncertainties as described in our SEC filings, which could cause future results to differ materially from those expressed in or implied by our comments.

Forward-looking statements in the press release that we issued yesterday, along with our comments today, are effective only today and will not be updated as actual events unfold.

In our discussion today, we will talk about 2019 results, excluding merger-related costs and reimbursed revenues and related expenses.

GAAP results appear on pages A-1 and A-2 of the earnings release, but our remarks today will largely refer to the adjusted results that appear on the non-GAAP reconciliation pages.

Of course, you can find our earnings release and reconciliations of all non-GAAP financial measures referred to in our remarks on our investor relations website.

As we begin our call this morning, it is obvious that the question you are most interested in is the impact of the coronavirus or COVID-19 on our business around the world.

In the 6 weeks or so that we have been intensely watching this crisis, we have learned much, but there's still a great deal we do not know.

In our press release and in our comments this morning, we will give you some yardsticks to help measure what the impact to our P&L might be.

While this is still guess work to some extent, we know one thing with confidence.

This will pass.

And when it does, the impact to our business will quickly fade.

So let's talk about our results.

We are pleased with our solid performance in 2019, finishing the year on a high note.

In the fourth quarter, we continued to add to our RevPAR index gains, increased hotel profit margins, recycled a meaningful amount of capital and signed a significant number of new hotel deals.

We grew our system to more than 7,300 properties and expanded our global rooms pipeline to a record of more than 0.5 million rooms, nearly 1.4 million rooms in 134 countries and territories.

We offer unrivaled choices for our customers.

In 2019, our development team signed 815 hotel agreements for a record 136,000 rooms with each of our international regions setting records in organic signings.

Over 220,000 of the rooms in our 515,000-room pipeline are already under construction.

Using 2019's pace of openings, our under-construction pipeline represents nearly 3 years of gross rooms growth, while our total pipeline represents well over 6 years of implied rooms growth.

At year-end, 7% of global industry rooms flew one of our flags, while our share of STR's worldwide under-construction pipeline led the industry at 19%.

To be sure, our signings were impressive, but we are not just focused on adding units.

We are focused on adding valuable hotels that drive higher fees per room and enhance our brands.

Luxury and upper upscale rooms comprise over half of our distribution globally, which is one reason our fees per room lead the industry.

During 2019, we expanded this lead by signing a record 45,000 rooms in these tiers.

At year-end, the number of our global luxury and upper upscale rooms under construction totaled more than the next 3 competitors combined, according to STR.

Other milestone achievements in 2019 included multiple launches from our new loyalty program, Marriott Bonvoy; to our new home rental program, Homes & Villas by Marriott International; to our all-inclusive platform, which was augmented by our recent acquisition of the Elegant Hotels Group in Barbados.

These expanded offerings and program enhancements provide meaningful value to our owners and guests, help to drive loyalty engagement and provide additional ways for members to earn and redeem points.

Homes & Villas provides the opportunity for our guests to stay at 7,500 premium and luxury rental homes in 200 locations around the world.

In the all-inclusive segment, our guests can currently choose from 10 resorts with 7 more projects in the pipeline.

In the fourth quarter, we launched our Eat Around Town offering, where Marriott Bonvoy members can earn points by dining at more than 11,000 restaurants in the U.S. We also introduced peak, off-peak redemption awards, providing members with better value when they redeem points on lower demand nights.

In addition to benefiting guests, the new award schedule helps owners fill more rooms by shifting demand from stronger periods to slower ones.

Finally, we are piloting a program in select international markets that lets local members earn and redeem points dining at our hotel restaurants.

The response from our members has been extremely positive.

Collectively, these efforts, coupled with the strength of our brands and our broad distribution, drove Marriott Bonvoy membership to over 141 million members at the end of January.

This powerful platform remains a key competitive advantage.

And in 2019, paid room revenues from loyalty members increased 11%.

Redemptions were also meaningfully higher as our Bonvoy travelers enjoyed the wide range of choices offered by the program.

Member share of occupied rooms topped 52% worldwide in 2019, up 250 basis points versus 2018 and reached 58% in North America, up 320 basis points year-over-year.

We also continue to see solid growth from our co-branded credit cards with sign-ups 12% higher year-over-year.

With an improved yield management approach and an increase in Bonvoy members, more of our guests booked direct in 2019.

Worldwide direct bookings, including group sales, rooms booked by our reservation centers and bookings made on our digital platforms represented approximately 3/4 of total room nights booked during the year.

Direct digital bookings alone represented 1/3 of room nights.

Mobile bookings, a component of direct digital bookings, were up a strong 64% over the year.

At the same time, the percentage of nights booked through OTAs declined year-over-year.

Guests' intent to recommend and staff service scores increased during 2019, thanks to the efforts of our outstanding associates.

We also saw impressive revenue share gains across our portfolio.

Overall, our global RevPAR index accelerated throughout the year, rising 240 basis points in the fourth quarter.

For the full year 2019, our global RevPAR index improved an impressive 200 basis points.

Each of our continents saw growth in index with meaningful gains from both legacy Marriott and legacy Starwood portfolios globally.

It is worth mentioning that we are particularly pleased with the progress we are making with the Sheraton brand.

Over the last 3 years, approximately 50% of Sheraton hotels worldwide have undergone, are undergoing or have committed to undergo renovation.

We sold the Sheraton Phoenix Downtown last month after purchasing it just 18 months earlier, and we signed a valuable long-term management agreement.

We are confident that the Sheraton Phoenix Downtown will serve as a showcase to encourage renovations at additional Sheraton hotels.

Before we discuss our 2020 outlook, let me talk a bit more about the coronavirus situation.

Clearly, this is a major humanitarian crisis, and our thoughts are with the many people impacted.

As the virus emerged in Wuhan earlier this year, our teams assisted guests and provided support for associates whose lives have been significantly disrupted.

I couldn't be prouder of our associates in the Asia Pacific region who have worked tirelessly.

We continue to waive cancellation fees for hotel stays through March 15 for guests with reservations at our hotels in Greater China and for guests from Greater China with reservations at Marriott destinations globally.

We began to see the impact of the coronavirus on our business in mid-January with occupancy declines gradually spreading from Wuhan to other markets in the Asia Pacific region.

In February, RevPAR at our hotels in Greater China declined almost 90% versus the same period last year.

At the end of 2019, we had 375 properties with roughly 122,000 rooms across Greater China, representing 9% of our total global rooms.

Around 90 of these properties are currently closed.

In the Asia Pacific region, outside Greater China, what we call APAC, February RevPAR declined roughly 25% year-over-year.

For APAC, we had 412 properties with 100,000 rooms at year-end 2019, representing 7% of our total worldwide rooms.

February RevPAR in the Asia Pacific region overall has been running down around 50% compared to February of last year.

Outbound travelers from China in 2019 made up less than 1% of room nights in our system outside of Asia Pacific, and around 0.5% of room nights in North America.

To date, apart from a handful of citywide event cancellations, we have not seen a significant impact on overall demand outside of the Asia Pacific region, though the situation, obviously, remains fluid.

Given the uncertainty surrounding the length and severity of the coronavirus situation, we cannot fully estimate the financial impact to our business at this time.

So in our press release and our remarks today, we are providing a base case first quarter and full year 2020 outlook that do not reflect any impact from the outbreak.

This base case reflects the 2020 outlook our teams had prepared as part of the company's budget process based on the pre-coronavirus environment, including hotel-by-hotel forecasts, group booking trends and expected supply growth.

Leeny will frame how first quarter results could be impacted by the coronavirus based on current trends.

Now let's start with our base case RevPAR outlook for 2020 not impacted by coronavirus.

On a global constant currency basis, we estimate global system-wide RevPAR in 2020 will increase 1% to 2% in the first quarter and will be flat to up 2% for the full year.

In North America, recent estimates for U.S. GDP growth point to a modestly slower pace of economic growth in 2020 with lodging demand forecasted to increase around 2%.

Industry supply growth is expected to also remain around 2% with upscale supply expected to grow 4%.

We expect leisure demand will continue to outpace business transient demand as there has yet to be a step-up in business investment levels.

Overall, this implies a continuation of low RevPAR growth in the U.S.

Our group sales organization in North America had a great fourth quarter in 2019 with bookings for all future periods up 5.5%.

With this strength, our group revenue on the books in North America for 2020 is up at a mid-single-digit rate.

We have completed roughly 80% of our corporate rate negotiations.

And while we can't predict corporate volumes, completed negotiated room rates are running up 1% to 2% for comparable accounts.

Our first quarter is off to a strong start with the benefit of easy comps in markets like Washington, D.C. and Hawaii as well as continuing RevPAR index gains.

We expect base case North America RevPAR will increase 1% to 2% for the first quarter.

For the full year, we expect it to be around the midpoint of the global range of flat to up 2%.

For the Asia Pacific region, we assume base case RevPAR growth 2% to 4% for 2020 with weak results in Hong Kong expected to continue for the first half of the year before lapping easier comps in the back half.

Again, this does not include any impact from the virus outbreak.

Base case RevPAR in Europe could grow 2% -- 2% to 4%, excuse me, for the year driven again by strong demand from U.S. travelers and strength in Eastern European markets.

For the Middle East and Africa region, we assume base case RevPAR could grow at a low single-digit rate in 2020.

We believe the region will benefit from higher RevPAR in Saudi Arabia, Qatar and Africa, somewhat offset by lower RevPAR in the UAE.

Continued new lodging supply in Dubai will likely challenge 2020 RevPAR growth in the UAE despite the Expo 2020 event that begins in the fourth quarter.

In the Caribbean and Latin America region, base case RevPAR could grow at a low single-digit rate for 2020, reflecting more modest economic growth and political uncertainty in some markets.

For 2020, we assume 5% to 5.25% net rooms growth, including deletions in the 1% to 1.5% range.

Preconstruction and construction delays persist around the world.

Again, our rooms growth assumption does not include any impact from the coronavirus situation.

Before I turn the call over to Leeny, I want to thank all our global associates for their continued hard work, especially those in the Asia Pacific region who have shown such empathy and skill, managing through this challenging time.

Our culture is distinctive and it is a real competitive advantage.

And I feel very fortunate to work with such purpose-driven and caring individuals.

On a personal note, I had surgery in November and the doctors were pleased with how it went.

As part of my treatment plan, I am undergoing a few months of post-surgery chemotherapy.

And while I am now fashionably bald, I feel really good.

I'm grateful I've been able to work throughout, and I want to thank all of you for your good wishes and support throughout this battle.

And now Leeny will walk through our financials in more detail.

Leeny?

Thank you, Arne.

Our fourth quarter adjusted diluted earnings per share grew 9% to $1.57, which was $0.11 ahead of the midpoint of our guidance of $1.44 to $1.47.

We picked up $0.03 of outperformance from fees, primarily due to better-than-expected incentive management fees in North America and $0.06 from a lower-than-expected tax rate due to higher windfall tax impact and other discrete items.

We also benefited from gains on the sale of 2 hotels in North America, which totaled $0.32.

These favorable items were partially offset by $0.26 from 2 asset impairments, $0.03 of greater-than-expected general and administrative expenses related to legal costs, bad debt and unfavorable foreign exchange and $0.01 from lower joint venture earnings.

Fourth quarter 2019 system-wide comparable global RevPAR rose 1.1% in constant dollars year-over-year.

North American RevPAR in the quarter increased nearly 1% with RevPAR among our full-service brands up 2.4%.

Leisure markets like Hawaii and Orlando showed notable strength.

Our RevPAR in the Asia Pacific region increased 0.3% in the fourth quarter.

RevPAR in Hong Kong declined 54% due to continued protests, while RevPAR in Mainland China increased 2.4%.

Excluding Hong Kong, RevPAR for the Asia Pacific region rose 3.5%, with strength in Singapore and India.

Our RevPAR in Europe rose 2.8% in the fourth quarter, benefiting from continued significant U.S. demand and robust loyalty program activity.

Eastern Europe was particularly strong due to increases in both rate and occ, while in Southern Europe, Italy, Spain and Portugal also saw healthy RevPAR increases.

Fourth quarter RevPAR in the Middle East and Africa region increased 2.8%, with strong growth in Riyadh and Mecca in Saudi Arabia.

Qatar also posted solid results despite the continued political tensions in the region.

RevPAR in the Caribbean and Latin America region rose 0.5% in the fourth quarter with strength in the Caribbean and Mexico partially offset by declines in Chile and Panama.

Our fourth quarter gross fee revenue increased 7% versus last year to $974 million due to room additions, higher REVPAR, improved net house profits at managed hotels in North America and Europe and continued strong growth in other franchise fees.

Depreciation, amortization and other expense increased to $179 million in the quarter.

We recognized a $15 million impairment charge associated with the sale of a North American hotel and a $99 million impairment charge related to a leased hotel in North America.

Our fourth quarter adjusted tax rate of 21% was higher than the prior year, largely due to favorable discrete items in the year-ago quarter.

For full year 2019, our gross fees grew 5% and our adjusted EBITDA increased 3%.

Excluding asset impairments and gains in 2018 and 2019, adjusted EPS grew 6% year-over-year to $5.92.

During the year, we returned $2.9 billion to shareholders, including $2.3 billion from share repurchases, thanks to successful asset recycling, strong cash flow generation and a reduction in cash balances.

Our loyalty program had net cash outflows in 2019.

This was primarily due to the marketing spend related to Bonvoy's launch in the first quarter and significantly higher redemptions as members explored the many new locations and experiences offered by the integrated and enhanced program.

We expect the Bonvoy program to continue to be a net user of cash in 2020 although meaningfully improved from 2019 levels.

We received proceeds from recycled assets of $470 million during 2019, including proceeds of roughly $310 million from the sale of the St.

Regis New York and $100 million from the sale of the Sheraton Gateway hotel in Toronto.

Now let's talk more about our base case outlook for 2020.

As you know, it does not include any impact from coronavirus, merger-related costs and charges, cost of reimbursed revenue or reimbursed expenses and it assumes no additional asset sales.

For full year 2020, given our assumptions for global RevPAR, our base case outlook shows gross fee revenue could increase 4% to 6% to reach $3.96 billion to $4.04 billion.

Growth should be driven primarily by room additions and other franchise fees, particularly offset -- partially offset by headwinds from renovations, property terminations and roughly $10 million of unfavorable foreign exchange.

Other franchise fees, which include credit card branding fees, hotel application and relicensing fees, time-share licensing fees and residential branding fees, are expected to grow roughly 10% to $630 million to $640 million.

We also expect that incentive fees will decline slightly given continued pressure on house profit margins.

We assumed owned, leased and other revenue net of direct expenses will total $295 million to $305 million in 2020, flat to up low single digits.

These results include slightly lower termination fees offset by a similar amount of favorable year-over-year impact from bought and sold hotels.

We assume general and administrative expenses will total $950 million to $960 million in 2020, up 1% to 2% versus 2019.

And we expect a 2020 effective tax rate of 23.3%.

We assume 2020 adjusted EBITDA will total roughly $3.7 billion to $3.8 billion or 3% to 6% over 2019 levels.

On the bottom line, we assume 2020 diluted EPS will be $6.30 to $6.53, up 6% to 10% versus $5.92, 2019's adjusted diluted EPS, excluding the impact of asset sale gains and impairments.

For first quarter 2020, our base case forecast assumes global RevPAR growth of 1% to 2% and a 5% to 6% increase in gross fee revenues to reach $940 million to $950 million.

Our tax rate in the first quarter is expected to be roughly 21%, 4 percentage points higher than a year ago as a result of higher windfall benefit and discrete items in the prior year quarter.

This translates to 5% to 7% growth in diluted earnings per share to $1.47 to $1.50 and 4% to 6% growth in adjusted EBITDA.

We remain disciplined in our approach to capital allocation.

Using the base case assumptions, 2020 investment spending could total $700 million to $800 million.

This includes around $200 million of maintenance investment spending, roughly $200 million of system investments that will largely be reimbursed by owners over time and $300 million to support new unit growth.

We expect roughly 3/4 of this new unit investment will be associated with luxury and upper upscale properties.

These projects typically provide higher fees per room and attractive 20-plus year agreements.

Projects where we invest our own capital are expected to generate a substantially higher value per key over the life of the contract on average compared to full-service deals with no Marriott capital.

Under our base case and assuming this level of investment, we would expect to return more than $2.4 billion of cash to shareholders through share repurchase and dividends for the full year 2020, assuming no impact from the coronavirus and no additional asset sales.

Note that this outlook assumes roughly $200 million higher cash tax payments than in 2019, primarily due to timing.

We remain committed to our strong investment-grade credit rating.

We ended the year within our 3.0 to 3.5x debt-to-EBITDAR target range, and our base case model assumes we will remain within this target range.

We will continue to evaluate the impact of the coronavirus situation on the company's cash flow and debt levels and manage leverage within our targeted range.

Turning back to the coronavirus situation.

Arne noted our distribution in the Asia Pacific region.

From a financial perspective, 2019 gross fees earned in the Asia Pacific region totaled $477 million, representing 12% of our global gross fee revenue.

Greater China generated about half of the fees in Asia Pacific, representing roughly 6% of both global fees and total adjusted EBITDA.

Our base case model assumes Asia Pacific and fees in 2020 will total roughly $500 million to $510 million, with Greater China fees again constituting about half of that amount.

Assuming the current low occupancy and RevPAR levels in the Asia Pacific region continue, we estimate the region will earn roughly $25 million less in fees and EBITDA per month as compared to our 2020 base case.

This means that for the first quarter, given our results in Asia Pacific to date and assuming the same low levels of RevPAR in March as we've seen in February and no meaningful impact outside of Asia Pacific, total gross fees and total adjusted EBITDA in the first quarter could be roughly $60 million below our base case and diluted EPS could be roughly $0.14 per share below our base case.

The analysis we are providing today has the benefit of actual results through the first 2 months of the quarter.

There's still a great deal we do not know, including the length and global scope of the virus and the impact of potential supply chain disruptions on the global economy.

As already noted, despite these unknowns, the virus will run its course.

And when it does, its impact will not be long-lasting.

We entered 2020 with tremendous competitive momentum in RevPAR, unit growth and brand strength and with an industry-leading loyalty program.

This momentum will carry us through this crisis and beyond.

We'll now open the line for questions.

(Operator Instructions)

(Operator Instructions) Your first question comes from the line of Shaun Kelley with Bank of America.

Thank you for all the commentary in some of the sensitivities.

I know this is a really kind of fluid dynamic, and I think we're all trying to get a hold of it for the global travel landscape.

So with that in mind, Leeny, as you think about some of the sensitivities you gave in the last section of your prepared remarks, are we really -- for that $25 million, are we really just extrapolating current trend line for, let's call it, Mainland China and Asia Pacific?

Or have we also accounted for the fact that this is spreading into, where we know so far, South Korea, Japan?

Appreciate that Western Europe is not probably part of that sensitivity.

But is there any -- so the question is, one, does it include broader APAC getting worse?

And then second, any sensitivities you could provide for us as we start to branch out into Western Europe, which we now know is an issue and then as we move kind of towards the U.S., which seems increasingly likely?

Thanks, Shaun.

Sure.

Let me start.

So the sensitivity we've given you is based on where we are in February, which is, as Arne described, is Asia Pacific RevPAR being down about 50%.

But obviously, that is massively skewed by Greater China being down 90% while the rest of Asia Pacific is down meaningfully less.

So that is based on an assumption that they stay roughly the same and that we continue to have no meaningful impact outside of Asia Pacific.

As you pointed out in your comment and your question, this is extremely fluid situation.

We are actually now reopening hotels in China every day.

But at the same time, how this exactly spreads to other continents remains to be seen.

Just in terms of the other continents, I think you're familiar with our basic layout of fees, which is that again, broadly speaking, you know that North America is roughly 2/3; Greater China, Asia Pacific, we described as being roughly 16%; CALA, 4%; Europe, 9%; and EMEA at 4%.

So all of these line up relatively well with our fee distribution as you look throughout the world.

The only other thing I'll mention, Shaun, is just to remember that from an IMF perspective, that Asia Pacific accounts for roughly 1/3 of our incentive management fees.

North America accounts for another 1/3 and the rest of international accounts for about 1/3.

So all of that fits into the equation that we gave you of the $25 million per month from Asia Pacific.

Your next question comes from the line of Robin Farley with UBS.

This is Arpine for Robin.

It sounds like unit growth of around 5% doesn't include any virus impact.

And you mentioned in your release that if the situation were to get worse, there will be impact to unit growth.

Is there any impact currently that's not included that you quantify?

And I know this could be challenging, but maybe you could provide some sensitivity similar to how you quantify fee impact in terms of unit growth.

It's -- I think it's too -- first, good morning.

It is too early to give you a numeric sensitivity to openings.

I think obviously, you've got, in Asia Pacific particularly, a very intense situation.

We did open about 1,000 rooms in January in China.

But in a sense, that's sort of before or at the very front end of this coronavirus.

I think if you look at the next few months, while we've got about 90 hotels closed and RevPAR down nearly 90% in the market, there's not much urgency to get a hotel open even if it's ready.

That if it's ready, though, it will open before the end of the year.

And so the impact on full year numbers may be nothing, but it wouldn't be surprising to see some of this get delayed.

I think in the rest of the world, it's much harder to assess.

We have talked to our design and construction folks.

We've talked to a number of our partners.

And I think, generally, it is decorative furnishings and some furniture and maybe some fabrics that are most likely to be sourced from China.

We think that the openings that were sort of planned for the first part of 2020 are more likely to have had all those supplies either in possession or en route to them and so could well not be impacted.

But I think we and many other industries, of course, are looking at the longer-term supply impact will be to the supplies that we need.

Obviously, in this context, it's more about hotels opening than the operating supplies.

But I think that there -- it wouldn't be at all surprising to us to see some further extension of the construction schedule in certainly -- as long as this virus situation lasts.

The only thing I'll add is to remember that new hotels opened throughout the year, and they're all ramping up starting from 0. So their fee contribution in year 1 is extremely small relative to overall fees.

Now obviously, year 2 is more important.

But the year 1 impact, frankly, from a bit lower openings, is not meaningful.

Your next question comes from the line of Harry Curtis with Instinet.

There are so many questions.

So many unknowable question -- or answers to questions.

So I -- maybe I'll focus on something that's a little bit more tangible, which would be the increase in your termination fees and your comments related to the renovations to the Sheraton brand.

Are the -- the increase in termination fees, were those more tied to the Sheraton brand or is it a mix of brands?

And do you see the -- what's the pipeline look like for kind of the legacy brands into 2021?

Sure.

Harry, we'll cover those.

So first of all, let's talk about termination fees.

Overall, termination fees in 2019 were meaningfully lower than they were in 2018, and we actually expect termination fees in 2020 to be even lower still.

So the ones in Q4, really, a question about timing and which hotels close and they can have a varying amount associated with them.

The other point I would mention, if you remember, last year, we had deleted rooms that started to get closer to 2%, while this year, we are squarely at 1% in terminations, which is on the lower end of our 1% to 1.5% guidance that we've given.

So I think from that standpoint, I think we feel good about the progression of how it's going with our portfolio.

I think in terms of the pipeline that we see, both in terms of legacy brands and in terms of the Starwood portfolio brands, I think, as we've described in Q4, we really kind of topped out a spectacular year in terms of new deal signings.

And they were happily very well distributed across all of our brands with some notable growth in some of the Starwood legacy brands.

Your next question comes from the line of Joe Greff with JPMorgan.

You touched on this in the press release and your earlier comments, Arne, about the solid RevPAR index growth both in the fourth quarter and for the full year.

Can you talk about how the Starwood legacy properties in North America performed, how much of the index fees would you attribute to those assets?

And then when you think about the performance in RevPAR growth this year, obviously, Greater China coronavirus neutral, how do you look at the Starwood legacy properties performing versus the Marriott legacy properties?

Yes.

The -- it's obviously a big world.

But as we mentioned in the comments, both legacy Marriott and legacy Starwood portfolios have been really performing extraordinarily well on index, both in Q4 and full year 2019 and as we start 2020 and both really hundreds of basis points.

The -- in Q4, there were some easy comparisons.

Obviously, we had a strike last year in the United States, which impacted San Francisco and Hawaii probably most.

They had probably a bit more impact on the legacy Starwood portfolio than on the legacy Marriott portfolio.

But even there, the RevPAR index performance in Q4 of last year was down a bit -- was down meaningfully less than it was up this year in Q4.

So whether you look at strikes or you look at a little bit of integration noise in Q4 of 2018, we not only made up that ground but we lapped it.

There are other sort of spectacular numbers.

You can see from our Q4, China RevPAR numbers, excluding Hong Kong at plus 2.5%.

I think the China team saw a RevPAR index growth for the 2 portfolios of plus 600 to 700 basis points.

And that is -- it's all cylinders moving.

It's the loyalty program, it is the digital platforms and the way they're performing, it is the sales team.

There's good news sort of across the portfolio, and it is very much shared by both the Marriott and the Starwood portfolios.

Your next question comes from the line of Jared Shojaian with Wolfe Research.

So maybe a question for Leeny.

If I look back at your operating cash prior to 2019, you were run rating about $2.4 billion before some of these cash headwinds that you've called out, particularly on the Bonvoy redemptions but also with the cash taxes.

So can you help me think about what you're expecting for 2020?

And does 2021 get back to sort of that prior run rate level that you can grow from.

And obviously, I realize that a lot of that depends on how long this coronavirus impact lasts.

But I guess what I'm ultimately getting at is, were there any benefits to pre-2019 cash flow that you just don't expect to see anymore?

So yes.

So a couple of things that I pointed out.

One is that we definitely had a bit of a benefit on the cash tax side in '19 that we will then pay for in 2020, relative, for example, on the cash taxes that we're paying on our asset sales.

So that is a bit of timing that will even out, obviously, over time.

On the loyalty side, I think that is the one that is worth spending a little time talking about.

And there, I think you definitely saw that in 2018, we saw the loyalty program behave more in its more historical pattern of being a cash -- net cash positive part of our story.

And this year, it moved to being several hundred million of a net cash user.

And that, you really need to think of within the context of the introduction of Bonvoy.

I think there was some pent-up demand relative to our customers being excited about being able to explore all of our properties and use their points at a much more expanded portfolio.

We also had the introduction of Bonvoy, which moves some timing of marketing expenses from 2018 to 2019.

And you put that together, and I think in the first year of the program, you definitely saw a fairly unusual pattern for the program.

We are quite confident that, that will smooth out over time and return to its more normal patterns.

We do think it will still be a user in 2020 but much less of a cash user.

And things that we've talked about, like the introduction of peak, off-peak, which manages the demand a little bit better in terms of the points that it costs at the different properties, both in low and high-demand times, all of that will work towards getting this program to where I think it behaves more like it has in the past.

Your next question comes from the line of Patrick Scholes with SunTrust.

Are you seeing any discernible uptick in cancellation activity outside of the Asia Pacific regions, at areas, say, like airport hotels or in general?

Yes.

Let's -- obviously, the weekend news around coronavirus was not good.

You had South Korea and Italy both and the Iran story as well a little bit.

Obviously, we don't have any in Iran, and so there's no measurable impact there.

But we're just days into it.

So we are essentially every day getting the team together by phone and getting data where we look at performance across these markets.

And we're listening to our customers, obviously, talk about it.

And let me give you a few anecdotes.

Maybe start with Asia Pacific, even though your question focuses on the rest of the world.

I'll talk about the rest of the world in a second.

China itself, Leeny mentioned about 90 hotels closed, we have RevPAR down about 90%.

I think the last full week number I had was minus 87% year-over-year, so not quite as bad as minus 90%.

The Chinese government is trying to ramp at least some things back up.

So we can see, for example, in Macau, we probably got down to 1% to 2% occupancy.

We may now be at 7% to 8% occupancy.

Now that's still down massively year-over-year.

I think it's too soon to put much stock in this effort to reopen China because it's early, and you still have schools closed, and we don't really know exactly how this is going to come back.

But there is at least some hope, I suppose, that we've bottomed in China and maybe things will get a little bit better.

You move around the rest of Asia Pacific and you see some things that you would expect.

So Singapore, down about 50% RevPAR year-over-year.

That's a, again, a recent week number.

That's not a full year February number.

By contrast, India, up 5%.

That was before President Trump's visit, so that's not driven by his visit but is driven by the fact that India is really not much dependent on China travel and has got a different GDP story than one which is dependent on the China story.

Obviously, when you look at South Korea and Italy, we will see both cancellations and we will see declining RevPAR in those markets.

Still too early to tell.

I know that some of the Italian cities, we've probably lost a few tens of points of occupancy in the first days.

But that's not the country as a whole, and it's far too soon to come up with sort of predictions for that, if you will.

I think when all is said and done, we would have to characterize our $25 million a month.

Run rate as being probably a bit light because we're going to see some impact in Europe.

We're going to see some impact in other markets around the world, which is probably not entirely dependent on China travel.

And our $25 million is basically a China travel story and an APAC story.

But I think even though that we would expect this will be messy for the next few weeks, if not maybe the next few months, we'd go back to what we've said before, and that is that this will end.

It's clear that it will end.

We can't tell you when it will end.

But when people start to get confidence that they don't need to be worried about picking this up if they're thinking about going to Seoul, for example, that travel will come back, and it will probably come back fairly quickly.

Your next question comes from the line of Smedes Rose from Citi.

I wanted to -- just wanted to ask you, you broke out, so your CapEx line item is about $300 million towards new unit growth.

How does that compare to '19?

And just are you just seeing more opportunities that you want to go after?

Or is that landscape becoming more competitive?

Or maybe just a little more color around that expenditure.

Sure.

So generally, I'd say, similar, maybe a tad higher relative to '19.

But again, it ties, as I said before, to the reality that these are generally on fantastic, full service and luxury projects that are well worth the investment.

And I think it ties to our success in signing new deals in these hotels around the world where the owners want our brands and generally, the market is competitive for capital for those projects.

But when we look at the value that we get on these hotels, we expect it to be meaningfully higher than the ones that require no capital.

So again, a little bit higher but not meaningfully.

Your next question comes from the line of Anthony Powell with Barclays.

Just a couple of questions on the loyalty program.

Question on the loyalty program.

You mentioned the positive impact of the redemption activity in the quarter.

How did points earnings trend in the quarter?

And are you happy with the level of activity around earning points in the system?

And given the positive impact of redemptions, does it make sense to run the loyalty program over more of a cash-neutral position over time rather than cash-positive position over time?

The -- a bunch of questions there.

I mean I think we talked about our penetration, too, which is both paid and redeemed nights as a percentage of total nights in the hotels, and we are seeing those penetration numbers move meaningfully up.

And to state the obvious, that means the program is growing enough to deal with the roughly 5% unit growth plus drive increased penetration in comp hotels, and we're very pleased with that.

So we're seeing both paid up about 10% or so and redeemed up significantly, too, and that's all gratifying for us.

I think -- well, it's a little harder to get share of wallet data because we guess on that a little bit.

We're obviously -- we don't have internally the kinds of tools we need to measure share of wallet.

We're quite convinced we are increasing share of wallet on -- from the loyalty program and from our loyalty members.

I think longer term, and Leeny's talked about both cash flow impacts in 2019 and in 2020, there's absolutely no reason that the loyalty program won't get back to being a positive cash flow generator for us on an annual basis.

That is, obviously, driven significantly by the fact that we continue to grow the program, and we continue to grow our portfolio of hotels.

And so as you do that, we will tend to issue more points for paid stays than are redeemed.

And we will continue to see, as we've done in the past, that there are more and more revenues coming into this program, which are coming in from credit card partners or restaurant partners or other partners besides just the hotels that are participating in the program.

So this is -- obviously, to be cash flow negative in the loyalty program in 2019 is unusual in a sense.

We'd love to wave a wand and have it be something different.

But it is actually quite logical given the launch of the program, both to market it, the new name and to get this massive group of customers to experiment with it.

And we are much more pleased than we are disappointed because it shows the engagement of our loyalty members with the program and with us.

Your next question comes from the line of Bill Crow with Raymond James.

You kind of classified the areas outside of China as not materially impacted.

I saw in the STR data yesterday that occupancy was down 100 basis points or more kind of everywhere, different segments and everything else.

I'm wondering if that's kind of the start of the impact that we should expect to see.

What are you seeing as far as real-time inbound international travel from areas outside of China and cancellations when you think about maybe the gateway markets?

I think the fair response, Bill, is we're asking the same question you're asking.

And what we get back at the moment is very much anecdotal, doesn't really show up in our data yet.

We obviously get our weekly flashes, and we get a daily look if we want to dig in and get a daily look.

And by and large, you look at the U.S. market, for example, which, just as a reminder, is basically 95% to 96% domestic travel.

So all business in the United States coming from international markets is in the 4% range, maybe 4.5%.

And big markets in that would include neighboring markets like Canada and Mexico, which probably have a different kind of travel profile, if you will, than the travel coming in from further abroad.

And there, we've got very, very few cases in the U.S. Obviously, we're all watching that to look at.

But we're not really seeing a measurable impact yet.

We're instead seeing, as we mentioned in the prepared remarks, a handful of group things really globally, which have canceled so far.

I suspect it will step up a little bit, but we're going to watch that on a day-to-day basis.

And overwhelmingly, obviously, that depends not just on time, but it depends on what are the incidences of the cases of this virus in various markets in the world and how do travelers react to that.

I appreciate that.

If I could just follow-up and ask if there are any takeaways from -- we went through SARS and the bird flu, and we've gone through a number of different coronavirus or some sort of flu-like thing over the last, what, 15 years.

Anything that you've taken away from that, that you can kind of guide your -- the company based on that?

Yes.

Leeny will have that precise number here that is worth talking about.

But the comparison to SARS, which is probably the most similar virus, is very hard to make.

I think if you go back to 2003, I'm guessing here a little bit, Chinese annual outbound travel would have been, I don't know, sub-10 million trips a year.

And last year, we were probably closer to 150 million China outbound trips.

So the relevance of China to the rest of the world is dramatically different.

The second thing, I think, if you look at Marriott's own story, we had -- we mentioned we've got 375 hotels open in China at the end of the year.

Those are not all comp, but our comp hotels are probably 2/3 of that or something like that, maybe a little bit more than that.

I think if you go back to 2003, we had 11 or 12 comp hotels.

They would have been mostly in Hong Kong, probably, and then a couple of cities in China.

And so I don't think there's much that we can really take from that other than when SARS ended, it ended and people got back fairly quickly.

That doesn't mean they get back the day after, but it does mean they get back within a month or 2 or 3.

Roughly a quarter.

Yes, pretty quickly.

But ultimately, that depends on people being able to look and say, "Yes, it looks like that's behind us." And so I think I think that's the comparison that's easiest to draw.

It's logical, not just SARS and MERS, but other unfortunate events that have occurred tell us that travelers are pretty resilient.

And one of the reason to be concerned is behind them, they're going to get back and get on the road.

Your next question comes from the line of David Katz with Jefferies.

Arne, great to hear you sounding well and I'm sure, looking fabulous.

I wanted to just touch on the IMF dynamic, to your point, that at some point, this will end.

How should we think about what happens to management contracts on the back end?

Are there embedded triggers within some of those contracts that move or slide that, that can actually help you earn IMFs faster on the back end of this?

How should we think about that?

I think the short answer is no.

I mean I think the -- it's a lovely thought.

But the way these incentive fees work basically -- and Leeny, correct me if I'm wrong here, but they're mostly annual tests.

The -- if there's an owner's priority, they're mostly fixed and they don't step up, which is a fabulous thing for us over the length of time.

But similarly, they don't mostly step down.

In fact, I don't know of a single contract in which an owner's priority would step down based on performance.

I think the -- probably the disappointing thing here is if you hypothesized that this was a 3-month issue in 2020 and of course, that's a total guess, but just use it for the sake of discussion.

The test is still an annual test.

And so the -- in the United States, particularly where we've got owners' priorities typically in managed hotels, the impact, if there is one, during those 3 months, will have some lasting impact on incentive fee earnings in 2020 but will have no impact in 2021.

I think it's the way to think about it.

If you go to Asia where typically, you don't have an owner's priority, we will not have probably the -- quite the hangover impact.

So if business disappears in China for 3 months and then it bounces back and does its sort of normal fee, if you will, we'll lose 1/4 of the incentive fees we would otherwise earn.

And so by the time you get to Q3 or Q4, we should be back to sort of a similar kind of pace as we've had in the past.

So the only thing I'll add to that, David, is just to remember the reality that internationally, we earn 80% of our hotels' earned incentive fees in 2019, while in North America, it was 56%.

So -- and these are very similar numbers to a year ago with a little bit lower number in North America because of the cost pressures and the low REVPAR.

And the other thing I'll mention is that it is the case in Asia Pacific that it's quite common that there is a slight increase in the amount of IMF as you increase your GOP margin.

So it is the case that as you get fuller and fuller and a really robust RevPAR that you can be earning an incentive fee that is instead of 6%, it becomes 8%.

And so there is the reality that as you are losing RevPAR, at first, it's a pretty dramatic drop.

But then obviously, the closer you get to 0, you're down at a lower level of earning IMF.

So there's less to lose.

But Arne's point is the right one, that these are annual tests.

So you got to really look at the end of the year, what you earned for the year and that will determine what you make.

Your next question comes from the line of Michael Bellisario with Baird.

Just kind of a two-parter.

First, on Avendra proceeds, how much is left there to be allocated?

And then kind of second, along the same lines on business interruption insurance, I know owners can get made whole.

But is it possible for you guys to recoup any loss income?

And then how might be Avendra proceeds be part of this to help you and your owners during this down period?

So let's just kind of -- as a quick refresher on Avendra, when we sold Avendra, there was a gain of, call it, $650 million that we're going to use to offset costs that otherwise would have been charged to the owner.

And I would say that we are roughly half the way through those monies.

And again, as we think about all the different programs and things that we're doing, those are obviously a part of what we would expect going forward, that we would continue to use to offset cost that otherwise would get charged.

But honestly, we do think of them as things that are kind of core to what we want to do for the hotel system.

I don't think of them as really ones that we kind of use it to plug a hole.

We've thought of them more as kind of ways to invest in the system.

And though, of course, it's great that we do have it and we can use it, I think at this point, we continue to expect that we will be able to use it to invest in the system.

And then on business interruption insurance, I think the right assumption here is that there will be relatively few policies that are implicated by the coronavirus.

We'll obviously watch that and make sure we study it.

But my guess is neither the owners or Marriott are going to get substantial business interruption proceeds from this.

In Asia Pacific, most of the hotels procure their own, and so it will depend on the reading of each of those contracts, of what their insurance policies say.

But we would not necessarily expect a big amount, at least in Asia Pacific.

Your next question comes from the line of Wes Golladay with RBC Capital.

I'm just hoping to learn a little bit more about the profile of the owner in Greater China.

Are they capitalized to the point where they can absorb a prolonged closing of their hotels?

It's a good question.

The -- of the 375 hotels that we had opened at year-end 2019, I know of only 1 that was not owned by a Chinese investor.

Those investors are -- cover the gamut.

I think there are many which are government-owned entities but not all.

There are a number that are substantial real estate companies that do own hotels but also do residential development in the rest.

And obviously, we've been in communication with owners continuously throughout the 6 or 7 weeks that we've been looking at this.

We have had really no indication yet that there are owners under severe pressure.

At the same time, when RevPAR is down 90%, it's a fair assumption that none of these hotels are producing positive cash flow to service debt or to do anything else.

I think one of the advantages of the economic system that China has is the government is involved not just sometimes through the ownership with government-owned entities but on the lending side as well.

And I think the government will have the tools in order to make sure that people will be able to navigate through this without foreclosures and without sort of significant long term consequences.

Your next question comes from the line of Kevin Kopelman with Cowen.

Great.

I just had a quick follow-up on the coronavirus impact.

So all the indications are that in the past week, the travel booking trends have gotten significantly worse.

Understanding that it's only 1 week, could you give us more insight on what you're seeing in the U.S., in particular, in terms of bookings for future states over the past week?

Yes.

And I -- again, as we mentioned before, we were picking up a few anecdotes, but we're not really picking up yet in the data anything that we can really, what, measure or predict from.

I think it is brand-new.

Obviously, you had, over the weekend, stories that we're focused on these other markets outside of China but also outside of the United States and Italy and South Korea and the like.

You've got the President for the first time speaking about it late last night.

And so the U.S.-focused discussion is obviously brand-new.

The numbers of cases in the United States are still tiny.

We were at a -- my wife and I were at a dinner in Washington last night with a bunch of folks who are obviously asking questions around this.

And we don't have a single case in the Greater Washington area.

And that's the case in most markets across the United States.

So I think it's way too early to expect that the data is going to be very revealing to us.

But at the same time, when you get the President doing a press conference on it, that's, by itself, not a good thing, and it will cause more travels to stop and think about it.

And again, it's one of the reasons I would say that the $25 million a month number that we've used as a yardstick is probably, at this point, a bit light, but we don't know what other number to give you.

That's really helpful.

And then if I could ask about something completely different.

The RPI improvements, what do you think -- could you call out kind of the key drivers there that are allowing you to see that RPI improvement year-over-year, what the outlook for that is going forward?

And then if there's any way you could give us, just the actual RPI level that you finished up 2019?

Yes.

The -- I mean loyalty -- the loyalty program is, I think, the thing that we would call out the most.

We talked about penetration, generally.

We gave you the -- both North America numbers and the global numbers for penetration, again, what percentage of total rooms is driven by both paid and redeemed nights.

It won't surprise you to know that the penetration numbers are higher in the U.S. just because our brand is better known.

We've been doing business here for longer.

And it's higher in select brand hotels than it would be in full-service hotels because you've got less group business and the hotels are overwhelmingly driven by transient business.

But even when you look at all those differences, you see very healthy increases in penetration outside the U.S. and in full-service hotels and in resort destinations, all of which goes back to the loyalty thing.

We're not going to publish today what our index is by brand or even by segment.

I will tell you that I get monthly a global index report.

And the cover of that report has got a chart that starts in 2014 and ends essentially with the current month that's being reported.

And the -- and that summary chart breaks it down by 3 segments, luxury being one segment, upper upscale being one segment and essentially upscale or the select brands being another segment.

And every one of those 3 is at an all-time high compared to that -- the other numbers on that 2014 chart, and they continue to go up.

Your next question comes from the line of Vince Ciepiel with Cleveland Research.

First, curious -- throughout the course of 2019, a number of players in travel alluded to changes that Google was making, which resulted in less free organic traffic, coming to their travel websites and a greater reliance on paid as Google changes up how they monetize things with their Google hotel ads product.

I was curious if you have seen any impact in your business at all from that in the second half of '19 and into '20.

We're watching what Google is doing very, very closely.

The -- certainly, if you look back over, I don't know, a 3- or 4-year period, we, too, have seen that the paid search volume coming out of Google has grown from what it was 3 or 4 years ago.

We didn't see a significant move in the second half of 2019 for us.

Now having said that, just as a reminder, we obviously don't like to pay for paid search if somebody is searching our brand.

So imagine, for -- let's take a discussion that somebody's gone on Google and they've searched Marriott New York, we don't love to pay for that.

And often, we don't -- usually, we don't pay for that because we think that person is looking for a Marriott hotel.

And they're going to end up finding our site, whether it's through Google or whether they come in through some other path because they're focused on it.

In the same way, if somebody goes on and searches Hyatt New York, we're also not likely to pay for that because that's a customer that is sadly, maybe, but nevertheless, focused on some other brand.

That is not necessarily the case in either instance for an OTA.

An OTA is going to also be selling Marriott rooms or Hyatt rooms or other rooms.

And if they can get somebody to come from Google to their site and collect commission associated with that, even if they're selling the room that, that customer seems to be looking at, that's going to go into their calculation in a way that's very different from ours.

And so I think it's a long-winded way of saying that the first impact of changes by Google in their strategy are much more relevant to the OTAs than they are to somebody like us.

But we are watching this very, very carefully.

We've obviously got a good partnership with Google, and there's a lot of volume that comes through our digital channels.

And we'll look for ways that we can use those tools to help us do what we want to do, which is particularly find customers who are not now in our loyalty ecosystem.

Find a way to bring them into our ecosystem.

And if we can do that in a way that is cost effective, we'll do it.

And unrelated follow-up, maybe I missed it.

I don't think I heard anything about election year, what impact that could be, what you've seen in the last couple and maybe how it relates to group versus leisure versus transient.

So the experience we've had over decades is that in Washington, during the Presidential election year, that tends to be bad for transient demand.

Having said that, the group bookings in Washington for 2020 are quite strong.

And so we actually think Washington will perform reasonably well.

The only other point to make is an obvious one, which is wherever the conventions are, they'll get a benefit from it.

Where the conventions didn't occur -- well, none of these conventions obviously occurred last year.

So those cities should help because there will be incremental demand.

But we've got lots of politicians staying in our hotels in these various markets and we're glad to have them all.

Your final question comes from the line of Stuart Gordon with Bernstein.

Stuart Gordon actually from Berenberg.

Just on the shareholder returns, how are you thinking just over how you'll phase these through the year given what's happening just now?

And just a quick follow-up, could you break out the other fees between credit card fees and the other fee components within that bucket?

Sure, absolutely.

So it's a careful balance.

We really look to maximize the returns to shareholders when we've got excess capital.

And we have a commitment to do that and we want to continue to do that.

At the same time, we have the same commitment to maintain our strong credit rating.

And from that standpoint, we need to be looking very carefully at both what's happening now and what we expect to be happening as we move forward through the year.

So we've given you our base case without any impact from corona, but -- the coronavirus.

But Stu, to your question, we are going to, of course, have to take into consideration real life and that will impact how we have to think about share repurchase.

But we're looking at everything we can on the cash side, on the expense side, on the fee side, et cetera, to continue to balance those 2 efforts.

But they -- absolutely, as we continue to see what happens, if you assume my numbers were correct relative to $60 million of impact on the EBITDA line in Q1, that will have an impact on share repurchase for the year.

And then as you think about the other question, which was on the credit card fees, yes, for the year, we had $410 million for credit card fees for the full year.

The rest, you probably remember that we have time-share branding fees that roughly approximate $100 million.

And then the rest is split between the residential branding fees as well as some other smaller categories and app and relicensing fees.

The way I would describe it going forward in 2020 is that group as a whole, we expect to grow up towards 10%.

When I think about the credit card component of that, I would call that kind of mid-single digits, maybe around 6-ish percent.

All right.

Thank you all very much.

We appreciate your time and interest this morning.

Get on the road, come stay with us.

This does conclude today's conference call, you may now disconnect your lines.