萬豪國際 (MAR) 2018 Q2 法說會逐字稿

Good morning.

My name is Jennifer, and I will be your conference operator today.

At this time, I would like to welcome everyone to the Marriott International's Second Quarter 2018 Earnings Conference Call.

(Operator Instructions) I will now turn the conference over to Mr. Arne Sorenson, President and Chief Executive Officer.

Please go ahead, sir.

Good morning, everyone.

Welcome to our Second Quarter 2018 Earnings Conference Call.

Joining me today are Leeny Oberg, Executive Vice President and Chief Financial Officer; Laura Paugh, Senior Vice President, Investor Relations; and Betsy Dahm, Senior Director, Investor Relations.

First, 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 last night along with our comments today are effective only today, August 7, 2018, and will not be updated as actual events unfold.

You can find slides for today's discussion on our website at www.marriott.com/investor or in our 8-K filing.

In our discussion today about the income statement, we will talk about results, excluding merger-related costs, reimbursed revenues and related expenses; the year-to-date net adjustment to the tax charge related to the U.S. Tax Cuts and Jobs Act of 2017; and the year-to-date adjustment to the Avendra gain.

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

So let's get started.

Our second quarter was terrific.

Adjusted earnings per share rose 56% to $1.73, with better-than-expected fee revenue, owned, leased results and G&A expenses.

Adjusted EBITDA increased 15% and cash returned to shareholders totaled nearly $1 billion.

As an encore to our first quarter strength, worldwide system-wide RevPAR rose 3.8% in the quarter, at the high end of our guidance.

In North America, we continued to see strength in leisure and corporate demand, particularly in the energy, retail and professional services sectors.

System-wide RevPAR in North America increased 3.1% in the second quarter, with RevPAR at luxury hotels up nearly 4%.

Transient RevPAR rose 2.5% in the quarter, with higher room rates associated with group compression.

Group RevPAR rose 4.5%, reflecting the shift in Easter holiday.

Group attendance trends improved, while food and beverage revenue rose nearly 5%.

Group pace for comp and noncomp hotels for 2018 is up at a mid-single-digit rate.

For comp hotels alone, pace is up at a low single-digit rate.

Given today's very high occupancy rates, we expect transient RevPAR will grow faster than group in the near term.

At the same time, year-to-date, our group revenue bookings for all future periods is up over 17%.

RevPAR growth exceeded 5% in the quarter in New Orleans, Orlando, South Florida, San Francisco and Houston and increased at a double-digit rate in Toronto and Vancouver.

RevPAR in New York rose nearly 4% with higher retail and special corporate demand, particularly from tech companies and higher international arrivals.

We expect 2018 RevPAR for North America system-wide comparable hotels will increase 2% to 3%.

For the first half of 2018, we reported RevPAR growth of 2.7%.

We believe RevPAR growth in the first half of 2018 would have been a bit slower, roughly 2.5%, if we adjust for last year's inauguration and the lingering impact of the 2017 hurricanes.

For the second half of 2018, we expect reported North American system-wide comparable RevPAR will grow 1.5% to 2%.

If we similarly adjust our second half RevPAR growth estimate for the midweek Independence Day and the tough comparison to last year's hurricanes, we believe RevPAR growth would again be roughly 2.5%.

In other words, our top line forecast is steady as she goes.

On the international front, the growth in demand continues to be impressive with price gains as well as occupancy improvements.

System-wide constant dollar RevPAR in our Asia-Pacific region increased 9% in the second quarter, while RevPAR in Greater China rose 10% on very strong retail demand.

The 2018 Beijing Motor Show increased occupancies in that city, while Hong Kong and Macau benefited from strong leisure demand, particularly from Mainland China.

RevPAR in Indonesia increased 17% as demand returned to Bali.

RevPAR in Japan increased 7% on strong retail demand, while in India, greater group business in gateway cities also increased RevPAR by 7%.

For the second half, we expect RevPAR in the Asia-Pacific region to continue to grow at a high single-digit rate, albeit a bit more modestly than in the first half.

In the Middle East and Africa region, system-wide constant dollar RevPAR declined 4% in the second quarter due to the earlier timing of Ramadan and continued political tension in parts of the region.

RevPAR in Africa rose nearly 7 -- or excuse me, nearly 9% with strength in Algeria, Nigeria and Egypt.

By the way, I just returned from a visit to our properties in Egypt and Algeria.

There is a lot of enthusiasm for Marriott there and a lot of room for future growth.

In the third quarter, we expect RevPAR in the MEA region will increase at a mid-single-digit rate, benefiting from the timing of Ramadan, while fourth quarter RevPAR is likely to be flat to modestly lower.

In Europe, the World Cup was played in Russia for the first time, and we saw considerable last-minute demand.

Congratulations to the French team on winning their second World Cup title.

Across Europe, system-wide constant dollar RevPAR rose 5% in the second quarter, driven by strong results in Russia, France and Turkey.

U.K. RevPAR was flattish, likely due to lingering concerns about Brexit.

RevPAR growth in Spain and Italy was weaker, as many Europeans shifted their spring holiday stays to other venues.

For the second half of the year, we believe RevPAR in Europe will continue to increase at a mid-single-digit rate.

While Italy is likely to remain weak, Spain will benefit from easier comparisons.

Paris should be strong, and we expect Turkey will continue to improve.

In the Caribbean and Latin American region, RevPAR rose 8% in the quarter.

RevPAR for hotels in the Caribbean increased at a low double-digit rate as they continue to benefit from lower industry supply following last year's hurricanes.

RevPAR in Mexico declined modestly, reflecting traveler concerns about security.

In the third quarter, we expect RevPAR in the region will increase at a mid-single-digit rate, driven by stronger results in Mexico on an easy comparison to last year's earthquake.

In the fourth quarter, RevPAR should moderate as hurricane-damaged hotels reopen and comparisons become tougher.

Worldwide, we expect third and fourth quarter RevPAR will increase 2.5% to 3% and 2018 RevPAR will improve roughly 3% to 4%.

While we're encouraged by 2018 results, we are even more excited about our longer-term prospects.

From a competitive viewpoint, we've never been better positioned.

We lead the industry in number of rooms and distribution, geographically and across chain scales.

Our brands are powerful.

Our already premium RevPAR index increased 120 basis points in the last 12 months and is likely to continue to climb.

Our 3 loyalty programs lead the industry and are getting better.

We expect to unify our loyalty programs on August 18.

While we will retain all 3 loyalty program brand names until next year, customers will experience them as 1 great program.

Guests will earn points faster, achieve elite status sooner and redeem points more easily and without blackout dates.

Also on August 18, we will introduce full portfolio inventory on Marriott's direct channels, including our websites and mobile apps, which will enable guests to conveniently search for any of our hotels on all of our websites and apps based on brand, tiers, distance, price, amenities, transportation, nearby attractions and points of interest.

With over 6,700 properties, we are positioned to take care of guests whether they are on a Midwest road trip, making a sales call in Johannesburg or enjoying a luxury resort getaway in the South Pacific.

This growing breadth of product and the growing number of earning and redemption opportunities increases the value of our loyalty program as guests don't need to look further than our properties virtually anywhere they may travel.

In addition to earning and redeeming points at our hotels, Marriott Moments offers guests local activities and experiences from sporting events and concerts to cooking and tennis lessons.

In the second quarter, Marriott Moments revenue nearly tripled from first quarter levels, enhancing returns to owners and increasing guest engagement and loyalty.

Loyalty programs make lodging demand sticky.

A loyalty program with significant luxury destinations and experiences is magnetic.

Dreams about luxury Hawaiian holidays are motivating, particularly for travelers who spend a meaningful part of their lives on the road.

In 2017, the 477 properties in our 7 luxury brands represented just 9% of our worlds -- of our rooms worldwide but 17% of our loyalty point redemptions.

Our luxury brands also contribute 19% of our property-based gross fee revenue, which could equate to nearly $600 million in 2018.

Like loyalty and luxury, our booking engines are also a significant competitive advantage.

Guests can book our hotels directly through our websites, apps, call centers, group sales offices or on property.

In 2017, 2/3 of our transient business was booked through one of our direct channels, with more than half of these direct bookings on our websites or apps.

We have been encouraging direct digital bookings using special loyalty member pricing and our next-generation yield management system focused on nights when occupancy rates are high.

Because of our efforts, revenue booked on our branded websites and mobile apps increased materially faster than our revenue booked through the OTAs in the second quarter.

For hotel owners, we continue to improve productivity as well as recognize synergy savings from the Starwood acquisition.

We recently reduced commission rates for intermediary group business, which should improve house profit margins for our best-in-class convention and resorts network hotels.

In the third quarter, most owners should see additional savings as we make further reductions in loyalty charge-out rates.

For hotels hosting loyalty redemptions, we have announced a new sliding scale reimbursement approach.

We believe this will reduce incentives for hotels to cut room rates at the last minute.

Finally, beginning in 2019, we plan on implementing a program services fee to bundle above-property charges, including reservations, sales and marketing, revenue management and mobile guest services.

Our program services fee should be simpler, more flexible and more predictable for owners, and we believe it will further reduce costs to most of our hotels.

And this leads to development.

According to STR, we continue to have the largest pipeline of rooms under development in the world, including more luxury and upper-upscale rooms than our next 3 competitors combined.

We opened a record 23,000 rooms during the second quarter, and our inventory of rooms under construction worldwide advanced to more than 213,000 rooms.

We added nearly 40,000 signed or approved rooms to the development pipeline in the second quarter.

At the same time, we also removed 14,000 rooms from the pipeline that had not made enough progress towards construction starts.

At quarter-end, our pipeline stood at roughly 466,000 rooms, a few thousand rooms higher than last quarter and roughly 25,000 rooms higher than at the end of the second quarter of 2017.

Worldwide, our pipeline emphasizes the higher-value lodging tiers as we begin at mid-scale and extend to luxury.

Our luxury brands alone represent 11% of our pipeline.

Of the tiers in which we play, our brands represent 30% of industry rooms under construction worldwide and nearly 45% of industry rooms under construction in North America.

To ensure we are driving real value, our developments -- our development team's success is measured both on meeting targets for net present value as well as number of rooms signed.

When we announced our intention to acquire Starwood in late 2015, we noted the driving force behind the transaction was growth.

The acquisition would provide an opportunity to create value by combining the distribution and strengths of Marriott and Starwood, enhancing our competitiveness in a quickly evolving marketplace.

With the significant accomplishments at Marriott since that announcement, we believe we are on our way to realizing that promise.

These accomplishments are due to the efforts of many people at Marriott.

With the upcoming launch of our combined loyalty program, I want to express our gratitude to the entire loyalty organization, our information technology team, our marketing group and everyone on property throughout the system who have been working toward this day.

Thank you.

So for more about the quarter, let's turn the call over to Leeny.

Thank you, Arne.

For the second quarter of 2018, adjusted diluted earnings per share totaled $1.73, 56% over the prior year quarter and $0.38 over the midpoint of our guidance.

Roughly $0.02 of the outperformance came from better-than-expected gross fee revenue; $0.02 came from better-than-expected performance on the owned, leased line, largely termination fees; $0.07 came from better-than-expected general and administrative expenses, including lower-than-expected profit-sharing contribution and favorable timing; $0.24 came from gains on the sale of assets, including the sale of a hotel in a joint venture; and $0.03 came from favorable discrete tax items.

In the second quarter, gross fee revenues totaled $951 million, a 12% increase year-over-year largely from unit growth, RevPAR gains and higher incentive and branding fees.

Credit card fees alone totaled $93 million compared to $59 million in the prior year, while other nonproperty fees, including timeshare fees and residential branding fees, declined 3% to $38 million.

With the weaker U.S. dollar, second quarter fee revenue benefited from $7 million favorable impact from foreign exchange, net of hedges.

Compared to our expectations, fee revenue outperformed by $11 million at the midpoint, largely due to better-than-expected performance at franchise hotels and stronger-than-expected incentive fees in North America and Europe.

Worldwide house profit margins for company-operated hotels improved 60 basis points in the second quarter on a 4.1% increase in managed hotel RevPAR.

Our teams around the world have done a fantastic job as we continue to recognize property-level cost savings from procurement, productivity and other merger synergies.

Owned, leased and other revenue net of expenses totaled $89 million in the second quarter, a 9% decline from the prior year.

Property dispositions reduced owned, leased results in the quarter by $21 million year-over-year, somewhat offset by higher termination fees and higher profits from owned and leased full-service hotels in North America and Europe.

Owned, leased and other revenue net of expenses was $9 million higher than expectations, largely due to higher termination fees.

General and administrative expenses declined 7% in the second quarter, largely due to synergies associated with the Starwood acquisition.

Second quarter G&A was $33 million better than guided.

We expected our profit-sharing match would total $25 million in the second quarter, following $35 million recognized in the first quarter.

The actual amount of second quarter profit-sharing match was $2 million as enrollments lagged expectations.

We expect the impact of additional profit-sharing enrollments and related expenditures will total $18 million in the second half of 2018 as we shift some of the previously anticipated spending to the back half of the year.

As a reminder, we currently estimate the total cost of this program is $110 million.

1/2 of it is being funded by our gain on the sale of Avendra, with $55 million impacting our G&A line in the full year 2018.

This expense will not recur in 2019.

Dispositions exceeded $400 million in the quarter and gains on asset sales totaled $119 million.

We sold the Sheraton and Westin hotels in Fiji, the Sheraton Montreal, the Chicago Tremont, our interest in a joint venture which owns the Royal Orchid in Thailand and the W Mexico City and our interest in a joint venture which owns land parcels in Italy.

Another of our joint ventures sold The Ritz-Carlton Toronto, and we recorded our share of the gain on the equities and earnings line on that transaction.

You may recall that in late 2016, we outlined a target of $1.5 billion of asset recycling by year-end '18.

With the completion of these asset sales this quarter, along with the sales of other hotels, joint venture interests and loan repayments in earlier periods, our asset recycling has surpassed this 2-year target, already reaching nearly $1.8 billion.

We've secured both long-term management agreements and, where needed, commitments for significant renovations for 9 of the 10 hotels sold since late 2016.

Equity and earnings totaled $21 million in the quarter and included a $10 million gain on the sale of The Ritz-Carlton Toronto Hotel.

This transaction also included a commitment for renovation of the property.

Net interest expense increased $14 million in the second quarter due to higher interest rates on our commercial paper balances, higher debt levels and lower interest income.

Second quarter adjusted EBITDA rose 15% to $939 million despite a $17 million negative impact from sold assets.

Looking ahead, we expect worldwide constant dollar system-wide RevPAR will increase 2.5% to 3% in each of the third and fourth quarters, yielding 3% to 4% RevPAR growth for the full year.

Nearly half of our 466,000-rooms pipeline is under construction.

Over half is outside North America and 41% is in the upper-upscale or luxury tiers.

While not shown on Slide 10, 45% of the rooms pipeline is company-managed, but less than 10% of the pipeline is subject to an owner priority return.

For the full year, we continue to expect gross openings to yield about 7% growth.

Our new signings remain robust and the number of rooms under construction has increased more than 20% since the second quarter of 2017.

And we continue to have more rooms under construction than any competitor.

For 2018, deletions are running close to 2%, a bit higher than typical.

While there are different stories for every hotel, we are completing workouts of Legacy-Starwood properties and are being more aggressive in addressing product quality issues.

We expect the pace of deletions will slow in 2019.

On a net basis, this yields roughly 5% net worldwide rooms growth in 2018.

Given our worldwide RevPAR and unit growth assumptions, we expect gross fee revenue for the third quarter will total $915 million to $935 million, an 11% to 13% increase over the prior year.

Our fee revenue estimates assumes nearly $5 million in favorable foreign exchange impact in the third quarter.

We expect owned, leased and other revenue net of direct expenses will total roughly $65 million in the third quarter, which reflects stronger hotel results and higher termination fees as well as a $23 million negative impact from sold hotels.

Our guidance assumes no further asset sales beyond those that have been completed.

G&A should total $235 million to $240 million in the third quarter, including roughly $10 million for additional contribution for profit sharing.

These assumptions yield $1.27 to $1.32 diluted earnings per share -- adjusted diluted earnings per share for the third quarter, 21% to 26% higher than our 2017 third quarter adjusted EPS of $1.05.

For the fourth quarter, we expect gross fee revenue will total $929 million to $944 million, including a roughly $10 million benefit from foreign exchange.

Results from owned, leased and other revenue net of direct expenses should total roughly $91 million in the fourth quarter, reflecting stronger hotel results, higher termination fees and a roughly $5 million benefit from our purchase of the Sheraton Grand Phoenix, offset by a $13 million negative impact from sold hotels.

G&A should total $236 million to $241 million in the fourth quarter, reflecting continued synergies associated with the merger as well as an $8 million profit-sharing match.

We expect our tax rate in the fourth quarter will benefit from some discrete items, which should lower the tax rate for that quarter to roughly 20%.

Fourth quarter adjusted diluted earnings per share should total $1.47 to $1.52, a 35% to 39% increase over 2017 fourth quarter adjusted EPS.

For the full year 2018, we believe gross fee revenue could total $3.64 billion to $3.675 billion, a 10% to 12% increase.

Incentive fees could increase at a roughly 10% rate, and credit card branding fees could total $360 million to $380 million for the year.

Our estimate of full year owned, leased and other revenue net of direct expenses includes stronger hotel results, higher termination fees and an $80 million negative impact from the sale of assets.

Our estimate for adjusted EBITDA assumes a $67 million negative impact from the sale of assets.

Full year 2018 G&A should total $935 million to $945 million for 2018.

Compared to our prior guidance, our G&A has decreased modestly for 2018, reflecting lower estimated profit-sharing contribution, somewhat offset by higher workload and costs associated with changes to accounting rules and higher compensation expense.

These assumptions yield $5.81 to $5.91 adjusted diluted earnings per share for 2018, a 38% to 40% increase over adjusted EPS of $4.21 in 2017.

We expect adjusted EBITDA will total $3.45 billion to $3.495 billion, a 10% to 12% increase over the 2017 adjusted EBITDA.

Investment spending for 2018 could total $800 million to $900 million, an increase of $200 million from our last forecast due to the $255 million purchase of the Sheraton Grand Phoenix, offset by reductions in other spending.

Maintenance and spending -- maintenance spending in 2018 should total $225 million.

Year-to-date, we've recycled more than $500 million of capital through asset sales and loan repayments.

We repurchased over 14 million shares from January 1 through yesterday for approximately $1.9 billion.

With the benefit of higher anticipated earnings and cash flow, we now expect to return over $3.1 billion to shareholders through share repurchase and dividends in 2018.

Our balance sheet remains in great shape.

At June 30, our debt levels were consistent with our targeted credit standard of 3x to 3.25x adjusted debt to adjusted EBITDAR.

We appreciate your interest in Marriott.

(Operator Instructions) We'll take your questions now.

(Operator Instructions) Your first question is from Harry Curtis with Instinet.

Most of the questions we've been getting surround the reduced NUG growth estimate.

And I wonder if you could give us more detail on a couple of things, which include, is -- first of all, is this, do you think, kind of a onetime surge in deletions?

How might -- how long might this last?

It sounds to me like you feel that this -- that the deletion rate normalizes next year, but to what degree do you have confidence in that?

How much visibility have you got there?

The -- let me maybe first say NUG, I think you're referring to net unit growth as opposed to another brand.

We do have a few brands but NUG is, I think, not one of them.

But you're right to raise the question obviously from our guidance a quarter ago, we have assumed a higher level of deletions this quarter than we did then.

And it's important, I think, that we all try to understand what's driving that.

I think the place I would start is not about deletions but is about development because I think the core question which should be addressed here is what does it say, if anything, about our owning partners' appetite to grow with us and to affiliate their hotels with us.

And we -- what we're seeing in the development side is a much more powerful indication of the strength of our brands than anything that the deletion data would tell you.

In Q2, for example, we signed -- we added 40,000 new rooms to the pipeline in newly approved deals or newly signed deals that had not been counted before, 241 hotels.

If you do the math, that's adding a hotel a little more frequently than once every 9 hours.

And that shows you that around the world, our owners are saying these are brands that we want to affiliate with, and we're prepared to put substantial capital behind that desire.

What's happened though in the last quarter or so is we've continued to make progress on a number of things which ultimately impact deletions.

And of course, each hotel is its own story.

The deletions in the second quarter were only about 18 hotels, but they are the most recent ones that left our system.

And what we see is about 20% of them, by definition older assets, are contract expirations.

And while contracts when they expire can often be renewed, the core issue there is usually does it make sense for additional capital to be invested in those hotels to bring them up to current standards?

Or are they sort of beyond that point from an economic perspective?

And typically, at expiration, obviously, it's a meaningful question.

And so we lost some hotels from that perspective.

About 30% of the rooms that came out in the second quarter, came out because of hurricanes or earthquakes.

And while they may come back into our system at some point in time, we looked at the circumstances of those hotels and thought it could be years before they reenter our system.

And rather than keep them in our unit count, let's take them out.

And the balance in the quarter, and I think to many -- in many respects, the increase in the deletion estimate for the year is driven by a mix of product quality issues.

They can be quite unique sometimes that, again -- but it's driven by the economics of each individual hotel.

And while we would like to keep hotels in the system if they could be brought up to standards, if they can't get the capital that's necessary in order for them to stay in the system, we just assume that they left.

Sometimes, obviously, we have a different point of view with our owners about the positioning of these assets, and we work through that and see if we can come to a resolution that makes the most sense.

I think the last thing I'd say here, Harry, and suggested in your question too, we do not think this is a new deletion rate that we're going to experience for years to come in the future.

I think this is a mix of pushing product quality issues.

I think it is a mix, to some extent, of dealing with some unresolved legacy workout issues that were within the Starwood portfolio when we closed the transaction.

Some of those deals have not been resolved, and we're working our way through them towards resolution.

And hopefully, we'll get back more to the 1% to 1.5% kind of deletion range that we've talked about the last couple of years for the years to come.

That's helpful.

Maybe just a little bit more history here.

When you acquired Starwood, the brand that seemed to be the most in need of attention was the Sheraton brand.

Perhaps you could help us by giving your perspective on a scale of 1 to 10, where do you believe the Sheraton brand is relative to where you want it to be within the next year or so.

We're making great progress on Sheraton, and we've, obviously, had lots of dialogue with our Sheraton owners around the world.

About a year ago, we settled on a sort of prototype for the regular guest room in the Sheraton brand.

In June of this year, we rolled out a new idea for the public space for the Sheraton brand, which we did at the lodging conference in New York 1st of June and have had tremendous response from our owners and franchisees to what we're doing with the brand.

When we look at the portfolio around the world, what we see is about 75% of the Sheraton portfolio is on its way towards meeting those brand standards.

Now that includes those that are already there as well as hotels which are scheduled for renovation or maybe even under renovation, leaving about 1/4, which we're in discussions with most of those owners to see if we can't get to a place where the renovation is going to be done to bring it up to brand standards.

And obviously, it's in that bucket that we have some deletions from the system, which we're certainly happy to take, because long term, we think we're going to strengthen the brand.

RevPAR index for the brand is now above fair share, which -- it's moved a bit since we closed.

I think it's got a significant movement ahead of it if we can deliver the kind of capital and reinvention of the brand that's necessary.

So I think we feel really good about the momentum we've got for plans for the hotels.

I think in terms of the customer experience, we've got to get more of these renovations actually completed and available to customers before they will start to see the average experience of the Sheratons move materially.

Your next question is from David Beckel with Bernstein Research.

I'll just ask a high-level question about RevPAR expectations this time relative to last quarter.

Obviously, a lot has happened geopolitically.

Are you hearing anything from executives that would cause you to be more concerned about the back half relative to last time we spoke?

Yes.

So good question, and we were very deliberate about using the phrase steady as she goes in the prepared remarks.

I think we could probably describe good news and bad news here.

So to -- for the optimists who are showing up at this call and saying, "Okay, we had a 4% GDP print in the United States in Q2.

That should drive faster RevPAR growth maybe even in Q2.

Maybe you should build that into your Q3 or Q4 numbers." We have not seen that higher GDP growth show up in higher demand either in our system or in the industry.

In fact, if you look at Smith Travel data, rooms sold in the United States, which is, obviously, demand, in Q1, demand was up 3%; in Q2, demand was up 3.1%.

So essentially identical.

And in many respects, our system very much shows the same thing.

When you adjust for the calendar anomalies, I think our first 2 quarters, one was 2.3% and one was 2.5% RevPAR growth so, essentially, identical numbers.

Similarly, and this is repeating what we said in the prepared remarks, but when we look at Q3 and Q4, if you're worried that the 1.5% to 2% U.S. system-wide number is concerning, don't.

It is not a sign of softening.

It is very much an impact of the calendar or comparison anomalies.

July 4 midweek, obviously, you've already heard about it from other companies in this space.

We've got some shifting holidays, but probably the most significant thing in the back half of the year is the RevPAR comparisons get tougher because of the strength of the hurricane recovery efforts in Houston and Florida, particularly last year.

And when you adjust for those things, what we are seeing in our guidance built in is about a 2.5% RevPAR growth through the balance of the year.

And so it's very much steady as she goes.

We see under that probably still somewhat greater strength in the leisure segment.

In the group space, the corporate group is stronger than the associate group, and the corporate transient is probably right in the middle.

That's very helpful.

And just as a quick follow-up to that, the group booking strength that you've seen and called out in your prepared remarks, how much of that relates to the reduction in commissions and folks wanting to try to book group activity in advance of that change?

That's a really important point to raise.

Our group bookings in the first quarter for all future periods were very, very strong.

One of the reasons, perhaps the most significant reason for that, was our reduced commission rates took effect April 1. And so with a number of group intermediaries, they were accelerating their efforts to make sure they were booking before lower commission rates impacted them.

I think, similarly, when you look across the industry, you've had a number of other companies decide that they would reduce group commissions too.

None of those group commission -- lower group commission levels are in place yet, and I think they will be rolling in for some other companies effective in the fall or the first of the year.

And so that will have a little bit of a dynamic to the way group business is booked.

Having said that, I think many of the end consumers for group business are maybe not ambivalent about what the group commission levels are, but they're going to be much more interested in where they should hold their meeting, both in terms of services and facilities, and we think we'll continue to compete very well in that space.

Your next question is from Smedes Rose with Citi.

I just wanted to ask you, with the loyalty program now on a path to be more fully integrated from the consumer perspective, it seems like the next phase of the Starwood acquisition maybe is going to be measured by what happens on the revenue side versus on the expense side, which, I guess, is kind of largely behind you now.

So how can you -- will you think about kind of communicating to the Street your gains presumably in market share or RevPAR index?

Or is that something that you can sort of provide now as a baseline so that we can measure or you can talk about in a year from now, kind of where that is?

Yes, the -- we love the optimism in your question, and we have the same optimism.

We think going to one set of channel platform, so that's the websites and the apps and the rest of it as well as one loyalty program, all of which happens August 18, will be a powerful positive for the system.

As we mentioned, we won't get to one name for the loyalty program until sometime early in 2019, but we will have one program as of the 18th of August.

That means, of course, that loyalty members can get credit for elite status or stays in the Marriott and Starwood Hotels.

Previously, you had to earn your status in one platform or the other.

It also means that points being earned and points being redeemed don't need to be transferred between accounts.

And maybe most powerfully, it means that customers will show up on our website and they will see the whole portfolio instead of having to toggle back and forth between 2 different portfolios.

We think all of those should drive increased share of wallet, greater strength in the loyalty program.

We'll be looking at measures like size of the loyalty program, number of members.

We'll be looking at contribution of the loyalty program to hotels, and we'll, of course, be looking at RevPAR index.

It is a very hard thing to predict what the upside's going to be, but we're optimistic that we will see greater strength from this stronger loyalty program, and we'll do our best to communicate with you about the actual results we achieve.

One other thing I'd mention.

I -- we don't think we're done on the cost side.

Leeny mentioned the 60 basis points margin growth in both international markets and the U.S. market in Q2.

That is, we think, very strong performance given the relatively modest RevPAR levels.

We've also got some further cost synergies, which we'll be rolling out to the hotels in the balance of 2018, and we know that we're going to deliver efficiencies into 2019 and 2020.

So we want very much to drive both top line and margin improvement for the portfolio of hotels in the next couple of years.

Okay.

That's helpful.

And then I just wanted to ask one question.

Some owners, the REITs have talked about poor sales bookings at formerly managed Starwood Hotels, transitions around the sales force.

And I'm just wondering, is that process largely finished now?

Or is there any kind of additional detail that you can add maybe from your guys' perspective versus what we've heard from owners?

Yes.

So we've seen a few of those comments and seen a couple of references to this in some of the early notes that were published.

And I'll try not to go on too long here, but I want to make sure that we talk about this a little bit.

The best indication, obviously, about the way hotels are performing is RevPAR index numbers.

And as we've talked about for the last few years, RevPAR index is a rollup of each hotel's performance against typically the 5 or 6 hotels that are most relevant in their competitive set.

Typically, those are going to be similarly positioned in the chain scales, in the same geographic market.

Sometimes, if they're in the luxury space or they're in the group space, the geography could be a little bit broader in order to get hotels that are similar.

But they're the hotels that have been picked by Marriott and by the owners to say, "Okay, this is the group that is most germane to assessing the relative performance of this hotel." And as we said in our prepared remarks, we have increased index about 120 basis points over the last 12 months, which with a portfolio of this size and the kind of work that's been underway on integration and all the other things that is underway here at Marriott, is fabulously strong results.

And again, the loyalty program has not been merged yet, so we think we've got upside ahead of us.

I think also, you can look at a little less precise data, but you can look at RevPAR growth for the Legacy-Marriott Hotels and the Legacy-Starwood Hotels.

I think it's interesting to note that the Legacy-Starwood Hotels are posting RevPAR growth numbers that are comparable, if not even a little bit higher, than the Legacy-Marriott Hotels.

One last thing before getting to the comments that were made by -- in some other companies.

We have a portfolio of hotels, about 70 hotels, if memory serves, that we describe as our convention resort network.

And they tend to be the big hotels.

They are the ones most reliant on the sales force because they are most reliant on group.

And when we look at the performance of those hotels, both Legacy-Marriott and Legacy-Starwood Hotels, we see RevPAR index up a full point in the last 3 months and a bit more than a full point in the last 12 months.

So that -- all of that would say there is nothing systemically that we see that would suggest the integration is sort of negatively impacting the system.

Now we've seen the comments made.

I won't go through each one of them and I'm not going to name which hotel even I'm talking about, but I thought it was interesting that one of the companies talked about a specific hotel and its relatively soft performance in Q2.

We know by looking backwards that, that hotel had weak group bookings on the books well before Marriott acquired Starwood for these periods in 2018.

And in fact, in the second quarter of 2018, that hotel had an increase in bookings for future periods of 38% compared to prior times.

And that's the first quarter when they had the new integrated sales force working for it.

And so again, each hotel is going to have a different story.

We're not saying for a second that there couldn't be circumstances in which there has been staffing implications to the integration that's been done or there've been distractions or there've been other issues, but what we see generally across the system is not an integration impact to the performance that we've had, but just the reverse, a remarkable strength in the midst of all the change which is underway.

The next question is from Felicia Hendrix with Barclays.

So just kind of along those lines, just in terms of the puts and takes of the RevPAR results, the North America RevPAR results in the second quarter, to what extent were they impacted?

What would have your RevPAR looked like if you wanted to kind of smooth out for the sales force transition?

And then also as you're thinking about your guidance for the third quarter, how much have you accounted for in that or maybe for just the second half?

Okay.

So I won't repeat everything I just said, Felicia, but...

No, I don't want you to.

We do not think there was any impact in Q2, and we do not think there is any impact in Q3 or Q4.

So to the extent anybody is thinking the Q3 or Q4 guidance numbers we put out there are somehow lower than they would be because of integration concerns, that is not the case.

Okay.

You were talking about -- I heard you talk about index and I heard all of that.

I just didn't know if there was a different way to translate it.

Okay, let's move on then.

Just -- I wanted to get -- just again on the unit -- the net unit growth because I was just -- in the quarter, there seemed to -- you gave guidance in April, so something happened in the quarter, right?

So just wondering how much of the greater-than-expected deletions was due to the Dubai hotels.

And then also looking at 2019, can we expect that you get back to that net unit growth of closer to 5.5% to 6%?

Yes.

So a few questions in there.

But interestingly and maybe a little bit ironically, the deletions we actually experienced in Q2 were half as many rooms as we experienced in Q1.

And so as much as we would -- it might be more concrete to be able to look at the higher deletion number and what actually happened in Q2.

The higher deletion guidance we've given you is really based on what we anticipate in Q3 and Q4.

The Dubai hotels that you referenced are not in the Q2 numbers because they actually were not deflagged until July 1, if I remember right, or July 31 -- let's see, we're at August 8 -- we're in August now, right?

So yes, end of July.

So that will be a Q3 number.

What we're anticipating in Q3 and Q4, and it's based on discussions underway as much as some decisions that have already been made and implemented, is that for a mix of reasons, we're going to see a bit higher deletions.

The 2 -- we've tried to call out and I maybe babbled on too long a little bit and clouded this, but product quality is certainly a piece of it and we are being tougher on that.

I think another point that we have called out is, there were a number of workout discussions that Starwood had not resolved, I think in part simply because of the pendency of our deal or the sale of that company.

And our teams have been working with owners to try and get those resolved.

And those discussions sometimes lead to hotels leaving the system as opposed to staying in.

The only last thing I'd say, I think, and I don't want to be pollyannish about this, but when you look at something like what happened in Dubai, we are confident that each of the 3 brands we have represented there will be replaced with product that we could be very proud of in that market in the fairly near term.

Okay, that's helpful.

But as we just -- for modeling purposes, are you more comfortable in 2019 with that 5 number or with where you've been more recently?

I -- we don't have a '19 number for you right now.

We, obviously, haven't built our '19 budget.

I do think that the deletions at 2% are likely to go back down into that 1% to 1.5% range, and you should not view 2% as being the sort of new expectation.

Your next question is from David Katz with Jefferies.

So we've -- I think we've covered the unit growth discussion fairly well.

But one of the issues I wanted to address is we've always looked at the Marriott brand and where it sits in the hierarchy with Sheraton and Westin and the degree to which those are either overlapping or bumping up against them -- bumping up against each other in the development community, can you just give us a little color around how that is progressing?

I know there was some repositioning around Marriott early on -- rather, Sheraton early on.

Yes.

I hope I understand your question, Dave.

And I mean, we've got, obviously, Marriott, Sheraton, Westin.

We've also got JW Marriott and Delta all in the upper-upscale full-service space.

Obviously, those brands come from 2 different legacy portfolios, and they were often competing head-to-head.

Now each of those brands would have said that they were better than the competitors' brands in the same space.

They're all now part of ours.

And if what you're getting at is how do we essentially target each of those brands to minimize the -- both customer confusion and, I guess, the overlap, that's what our brand teams are hard at work at.

And they are hard at work at it with our owner and franchise partners.

Sheraton is the one we've talked about the most because Sheraton has needed it the most.

And I think what we're seeing is good buy-in from our owners and franchisees to move the average quality of the Sheraton portfolio up meaningfully from what it was when we closed on the acquisition of Starwood.

It probably on average will be a fraction lower than where the core Marriott brand is.

But that -- stress the word -- the words on average, because depending on the market, we will have one that's positioned a little bit differently than another.

The, obviously, by definition, full-service hotels have been opened and developed over many decades and location will still feel quite important to this.

And that's a comment that could easily be applied to the brands that have got a bit more of a lifestyle flavor too.

We've got Renaissance and Meridien that are both also in this full-service space, and they are each been positioned true to the heritage that they've had.

And I -- again, I think we're making really good progress on it.

And you can look at this and say it's a monumental task to get these brands with some distinctions between them.

But we can also look at them and say it's an extraordinary opportunity to have this kind of distribution in an economically significant space and a space which is really important for our loyalty members and our group customers to have some choice and to have the kind of product and services that we can deliver across this portfolio.

So we wouldn't want to be without any single one of these brands.

Your next question is from Patrick Scholes with SunTrust.

Just a quick question.

On the previous earnings call, you had noted that your group revenue-booking pace for luxury and upper-upscales was up 1% to 2% for the second half of this year.

What does that comparable figure stand at today?

It's about the same.

Unchanged?

Okay.

Yes.

Your next question is from Robin Farley with UBS.

Two questions.

One is just on the issue of the unit growth.

Just looking at it in the slides where you show conversions as a percent of your pipeline, it shows about 2%.

And one of your other large competitors is talking about conversions being about 20% of their pipeline.

I don't know if you and they are sort of measuring it in different ways, but that's such a big difference.

I don't know if you have any thoughts on why your conversion assets -- are you not capturing conversions or something?

Or are you just measuring it differently in these 2 metrics?

Sure.

I think certainly definitionally, I'm not going to pretend to know how they're included.

As you know, conversions can wander into your pipeline and stay there for 1.5 years if there's a meaningful amount of work that needs to be done before it opens or it can literally flip overnight.

So we're still in the land where we think a 15% to 20% of our room openings come from conversions, which has been consistent over time that, that's the way it's been.

When you think about our Autograph brand as a terrific example and our other soft brands, just terrific pipeline for us.

But again, in those situations, they very often don't enter into the pipeline.

Okay.

No, that's helpful.

And then my other question was just when you look at guidance for the second half, adjusting for what you've previously quantified as the impact of holiday shifts and the impact of the sort of hurricane-driven demand, it looks like even though you're giving the same RevPAR guidance for Q3 and Q4, if you back out the comps from the prior year, that you're maybe expecting an acceleration in underlying demand in Q4.

And I don't know if you could just give any color around what gives you that or what do you think would be driving that.

I don't -- I wouldn't interpret what we've said as expecting an acceleration.

What we're expecting is very much steady as she goes.

The -- adjusted for calendar and comparison issues, we see RevPAR for the first 2 quarters and the last 2 quarters of 2018 as essentially equivalent at about 2.5%.

I guess, I'm just thinking about Q4 versus Q3.

So but definitely understand what you're saying about second half versus first half.

But if, I guess, the impact of hurricane-driven demand and holiday shifts were greater in Q4, so if we back that out for Q4 to be at the same rate as Q3, seems like your underlying...

Well, you've got the July 4 timing in Q3, and there may be some -- I can't remember where the Jewish holidays are hitting and exactly how they're hitting.

I know our pinpoint calculation for Q3 and Q4 are 0.2 points apart, which, to be fair, is a smaller gap than our accuracy.

And believe it or not, there can be also quarter-to-quarter a day of week shift just in terms of how many Tuesday nights there are versus how many Sunday nights there are, and that also is a little bit of a factor between Q3 and Q4.

Your next question is from Bill Crow with Raymond James.

Arne, I think you mentioned in the formal remarks that part of the pipeline issue was -- or just something that weighed on the pipeline was taking hotels out of the pipeline that had yet to commence construction in a timely manner.

And I'm just -- I recall back to the days when a lot of Starwood's peers were suspicious about the actual strength of their pipeline.

I'm just wondering whether you've culled the -- fully culled the Starwood portfolio and pulled out any of the maybe phantom deals that were in there.

Yes.

I don't -- and I wouldn't attribute that mostly to a Starwood -- Legacy-Starwood portfolio.

I think that the -- what?

14,000 rooms we deleted, I think, in the quarter, we think we should pull that out, 13,000 or 14,000 that we pulled out of the pipeline.

And our team culls through that.

I won't say necessarily that every quarter is as intense as the last one.

But we were pretty aggressive in this quarter in going through and found, in both the Legacy-Starwood brands and Legacy-Marriott brands, that there were projects that had not moved fast enough.

And so we pulled them out of the portfolio.

And it does a few -- in some respects, this is not -- it's a little bit bigger number than a typical quarter would be, maybe meaningfully bigger.

But it's not a totally unusual thing to discover that the best-laid plans maybe are not coming to fruition with an owner someplace.

And I think in the environment that we're in today, we've got still plenty of available capital to invest in new projects.

Availability of debt probably is still plentiful, but equity requirements are maybe a little bit higher than they were before.

And construction costs are a little bit higher than they were before.

And so you end up with some folks who maybe signed that deal a year or 2 ago, and they haven't moved forward.

And ultimately, they tell us they're not planning to move forward on it.

We'd just as soon cull it out so that we've got that market to pursue with somebody else.

Okay, great.

And then the follow-up, I guess, NH Hotels appears to be in play.

I'm not going to ask you specifically about that.

But just your thought now that the -- you've made such progress on the integration of Starwood.

Are you ready to get back into the consolidation theme?

Or do you think Marriott's set for a while?

We will continue to look at opportunities that are available.

I think as it relates to NH in particular, what looks like Hyatt discovered is Bill Heinecke has that company reasonably well tied up.

So I don't think there's much point in talking about that one, as if we would anyway.

But we'll see.

We're obviously appreciative of the way the Starwood deal has gone so far.

We're also very appreciative of the way the organic growth story is going.

But as we sit here today, our share of the global hotel business is still not particularly huge.

And so I think there's plenty of opportunity to continue to grow.

Your next question is from Joe Greff with JPMorgan.

Two questions for you on the deletions -- room deletions commentary.

Relative to a quarter ago, what is the amount of forgone fees relative to that incremental deletion amount?

And if you could break that out between the workout category and, say, the other category.

I would imagine the incremental fees or lost fees, forgone fees is relatively low, but that could be [prospective put in].

Leeny, we have something specific on this, but let me just say, on the short term, a hotel leaving the system, we may lose management or franchise fees, we may gain in the short term termination fees.

And so when you look at it, for example, in 2018, I'm not sure there's much impact.

You look at it longer term, obviously, there is impact because those fees have left the system.

I don't know, Leeny, if you've got the numbers.

So a couple of numbers for you, Joe, and we certainly -- kind of diving down into how much is from workouts versus product, we don't have that level of detail.

And as you know, there are a whole host of different reasons.

But let's just kind of talk broadly.

Termination fees have generally run, call it, $15 million to $20 million a year.

I think this year, we are likely to be as much as double that, maybe even a little bit more, based on the numbers that you heard us talk about today.

So from that perspective, you're clearly getting the extra kick.

In the back half of the year, we did reduce our fees by close to $5 million as a result of the terminations that we expect to have this year.

So that is impacting this year's earnings.

When you then think about full year '19, just to give you a broad sense of what the full year terminations could impact.

If you're talking about, again, call it, very roughly 2%, then you can imagine that it could impact hotel fees next year by a bit less than that because some of these hotels aren't necessarily giving a lot of fees.

So somewhere between 1% to 2% of hotel fees that will be lost as a result of 2% terminations this year.

Understood.

And then on the group pace for this year, is there a huge difference or meaningful difference between Starwood legacy properties and Marriott legacy properties?

No.

The numbers are very close.

Your next question is from Thomas Allen with Morgan Stanley.

Two questions for me.

First on G&A.

You highlighted the Starwood synergies a couple of times.

How is that tracking versus your $250 million guidance?

Sure.

So if you kind of take the 2015 full year combined company adjusted for inflation and then using our current expectation for the full year of 2018 and adjust out for the $55 million, it's pretty much close to on the button in terms of hitting the $250 million.

And I guess, a follow-up to that is, do we think there could be incremental G&A synergies?

And then my second question, just to throw it in now, you touched on how you've outperformed your asset sale goal.

You still own some Legacy-Starwood Hotels.

How should we think about the plans with what's left?

Sure.

All right, let me do the second one first, and then I'll talk about the synergies.

Today, we own 14 hotels.

They're split half-and-half between Legacy-MI and Legacy-Starwood.

We are, as you know, still completely devoted to our asset-light model and continue to be engaged in selling those hotels.

They are, as you might imagine, not predictable in terms of the timing of those.

So that's why, as usual, we leave those out of any guidance going forward, but we certainly are still going at it.

On the G&A synergy side, I think the low-hanging fruit has definitely been taken.

When we look forward, if you look at a global economy continuing to stay strong, you clearly -- around the world, you've got some wage pressure that you have to take into consideration when you look at kind of how it might go in 2019.

Certainly, as I talked about on the last quarter's call, when you take out the $55 million -- when you include the $55 million this year, you can imagine that next year's G&A is lower than this year's printed number.

But you also -- kind of finding other big chunks of synergies, I think, is probably not realistic, but I think continuing to be more and more efficient is.

So we will continue to go after that.

Your next question is from Shaun Kelley with Bank of America.

I think a lot of everything has been covered here, so just one high-level one, but I do think it's something that's come up some.

So there's been a lot of press reports recently going into the August 18 kind of combination on the loyalty program side.

So Arne, could you just talk at maybe at high level of kind of what you're doing to, I think, keep the highest-level SPG members kind of pleased and happy through this whole process?

I mean, I know that it's a very difficult balancing act, but I would say that from the consumer side -- and obviously, you have to think about multiple constituents here, but from the consumer side, that was a kind of a core asset of Starwood when you acquired them.

So just kind of what are you doing to keep that kind of customer happy throughout the program and throughout the transition?

Yes.

So it's a good question.

It's obviously something we've been focused on, not just since we closed the deal but since we announced the deal.

And we -- the moment we announced in the fall of '15, we heard loud and clear from the SPG loyalists, "We love this program and make sure you protect it for us." And it has been a steady set of decisions and conversations since then, which have been aimed very much at doing that.

We -- even before we closed on Starwood, we did a couple of things to the Marriott Rewards program to send the message to the SPG loyalists that we were going to protect some of their benefits.

And they included things like late checkout, which had not existed on the Marriott Rewards side before, but we said rather than simply say it to the SPG folks that we're going to do it, they would believe it more if we actually did it on the Marriott Rewards side.

And there were other similar decisions that we made in that sort of time frame.

I think when we got to the day of closing, we also did a number of things.

I think one of the most important was the 3:1 points conversion ratio between SPG and Marriott Rewards, which I think SPG members looked at and said, "You know, that's a very fair conversion and it does well for us, and so we appreciate that." And then I think in the last, what, 9 months or so probably most intensively, we have focused on the economics of the program, which very much depend also on the economics of the credit cards.

And with better credit card deals, we have found that we can deliver great value to the elite members and the basic members of both loyalty programs.

We can deliver it at reduced expense to the hotels, therefore benefiting the hotel owners and franchisees.

And Marriott, obviously, can experience an increase in the credit card contribution to its own P&L.

And while it would be too much to say that every single Marriott Rewards or SPG member has stood up and applauded, I think what we've heard from the bulk of that community is you've made a collection of decisions that caused us to feel very good about that program.

And that's what we've intended, and I think we'll prove it in the way that ultimately it is used by our well over 100 million members and growing.

And maybe just to stay on the same -- the kind of the same idea, like at the very highest kind of levels, the Ambassador program for Starwood, take that, for instance, is the value of that guest as, like when you guys think about it from your side, as high as maybe some of these kind of the super frequent people?

Like is there some -- I guess, is it -- it's hard to articulate, but is there some exceptional value to that guest?

Or is everything sort of long and extreme because they also cost more to service and other things?

So how are you kind of balancing just that ultimate kind of elite member?

Because I think when we read some of the reviews and the blogs and some of the things like this and some of the articles that have been written, that's where probably the most discussion about the program comes up.

Yes.

No, we love those elite folks, and the Ambassador program we've expanded into the Marriott side of the equation too because we think it is exactly the right kind of step for these most valuable customers.

And so we're doing that.

And again, take -- you take this a day at a time in some ways, and we won't really have proven it until we get to one program and have it working.

But I think the decisions that have been made are exactly the right ones.

And I think the community as a whole is responding extraordinarily well to it.

Your next question is from Jared Shojaian with Wolfe Research.

So I want to ask you, you called out the $67 million impact for the asset sales in 2018, but can you give us what the number would look like for 2019, just based on the sales you've done so far this year?

And I'm assuming that's a net number that includes the Sheraton and Phoenix purchases, is that right?

So first of all, I'll say we haven't done our budgets yet for 2019, so it would be interesting to give you an expectation for the actual -- the lost revenue overall for owned, leased.

But on the hotels that we're selling this year, we'll -- we can get you that number.

I don't have the total here.

We'll work on it and get it to you.

Okay.

And that is a net number?

It does include the Sheraton and Phoenix?

Yes, no.

Okay.

And then I just want to ask, switching gears on China.

I think your Asia-Pacific RevPAR guidance is still pretty strong.

But maybe you can just comment on if you're seeing anything in China.

I know you're talking about steady as she goes throughout the entire business.

But maybe if you could just comment on what you're seeing, especially in light of all the trade talk right now.

Yes, obviously, we're all listening to the trade talk, I think, and it is anxiety producing in many respects.

I think what we've seen in terms of trading conditions for open hotels in China has been very comforting, performing well.

There's no sign that the trade conversation is impacting performance.

And what we've seen on the development side has equally been reassuring in that -- and part undoubtedly, because of strong hotel performance, for existing and open hotels, there continues to be a development appetite for Chinese real estate investors.

Remember, our business is substantially Chinese in that -- of the hotels we have opened in China, I can think of one that's not owned by a Chinese company.

And so the book of the economics are very much driven for the benefit of Chinese investors, real estate investors.

They are often government-affiliated companies, but not always.

And as a consequence, we may be in a somewhat different place in a trade kind of conversation than the typical industry.

I think the -- personally, my larger fear about the trade war potential is what it could do to GDP growth in the United States, and to some extent, what it could do to the cost for construction materials in the United States and in other markets around the world.

And that certainly has not seemed to manifest itself yet in U.S. GDP numbers.

It probably is starting to manifest itself in terms of some materials that are used for construction.

But it's early on in that process, and we'll have to see how it evolves.

Your next question is from Rich Hightower with Evercore ISI.

So I want to hit back on the Starwood sales force integration issues from the last quarter.

We've heard from at least a handful of owners, so not really just 1 or 2, but at least a handful, and we have yet to hear from your largest owner, of course.

So that's (inaudible).

Yes.

That's tomorrow presumably, right?

Yes.

Exactly.

So at the same time, I mean, everybody has got their own version of sort of what may have led to weakness of the individual hotels during the second quarter or whichever period we're talking about.

I guess, if you had to fairly articulate the view of those owners with respect to this issue, what do you think a fair representation really is just so we kind of get to the -- get clarity on this point?

Well, I mean, you've heard from them directly, obviously.

And I'd be -- be inappropriate for me to speak for all of them because I think their views are not monolithic.

Interested to note that one of the companies we heard talk about this in their public call had never raised the performance of the hotel that they raised in the call with us.

But -- so as a consequence, our team hadn't really had a chance to engage in a dialogue with them.

I think generally, though, and even what we've seen in the conversations over the last week or 2, generally, you see even in those comments that the impact is viewed as transitory.

They remain supportive of the transaction.

They remain supportive of the notion that this transaction is going to deliver a significant value to them from top line and bottom line synergies.

And even I think more of this concern has been raised with respect to Legacy-Starwood Hotels and I think a number of those folks have said, "We still believe that the Legacy-Starwood Hotels may benefit disproportionately from Marriott and Starwood coming together." And so we don't hear this as -- obviously, we hear something.

We hear the words that are being used, and we don't want to say that there can't be instances in which there is some impact or there shouldn't be conversations to make sure we understand this as well as we possibly can understand it.

But when you look at the system as a whole, we do not see it showing up in our data.

We don't see it in the RevPAR index numbers, we don't see it in the RevPAR year-over-year growth numbers.

We don't see it in what the customers are telling us about the way they're booking.

And we'll do everything within our power to make sure that, that continues to be the case.

And then in the meantime, we'll deal with the owners who called out concerns with respect to particular hotels and say, "Let's make sure we understand them and make whatever decision needs to be made in order to have those hotels perform as well as they can perform."

Your next question is from Carlo Santarelli with Deutsche Bank.

I just have one quick one.

As it pertains to the deletions that you talked about, if you go back to your March of 2017 Analyst Day and kind of frame maybe the change in the magnitude of the deletions from what your expectations were at that point relative to kind of what you're currently seeing, is that possible?

Yes, I'm not sure it is.

I think we can say that we've been talking about 1% to 1.5% deletions for a while.

I think actually last year, we were closer to 1% than 1.5%.

And so in a sense, we were positively surprised and maybe this is a little bit of a reversion to the mean in 2018.

But I'm not sure we can say much more than what we've already said.

I think 1% to 1.5% for now and, again, we will get to a point where we share a multiyear plan with you all again at some point here, but until we get to next year's budget or until we get to a analyst conference, I think 1% to 1.5% still should be your expectation for the foreseeable future.

Great.

And if I could just ask one quick follow-up.

On the group pace for all future periods, you stated plus 17%.

Was that -- that's the comparable number, is that right, the 17%?

No, no, no, that is group bookings year-to-date for all future periods.

Bookings done in '18 for the first 6 months.

That's not pace for all bookings from all time for future periods.

Your next question is from Vince Ciepiel with Cleveland Research.

I wanted to touch on something.

I don't know if it's come up today, but the Tribute portfolio for booking homes in London, I think as of the last call, you were only a few weeks in.

Just curious now that have -- you have a few more months, what are your early learnings there?

And then would it surprise you to have other cities outside of London on that platform maybe by this time next year?

Yes.

So thanks for raising it.

Just as a reminder for everybody, we launched about 90 to 120 days ago maybe a pilot in London under the name Tribute Homes.

It is a relatively small pilot in the sense that there are about 200 homes that are connected to our system.

And they are bigger than studio apartments.

They're bigger than hotel rooms.

They're whole home products.

So that they're -- it's a product which feels different to us from both what we have in the traditional hotel space and what the typical product is that a number of the home-sharing companies are focused on.

And we wanted to provide loyalty linkage, and we wanted to provide a set of services that would make it more predictable and more consistent with sort of the brand promise, if you will.

So not just key delivery services, but housekeeping services and design and decor services, which are delivered by our partner in London.

And it, we think, it allows us to distinguish a little bit both in terms of size and in quality from sort of the average home-sharing theme.

Small pilot though.

The first -- this -- so far, it's gone great.

Our loyalty customers seem to like it, not surprisingly.

It is predominantly a leisure buy.

Not surprisingly because it's a leisure buy -- it's only London, obviously, it is a bit longer stay than we anticipated.

So the stay we're experiencing is closer to 5 nights on average, which is a little bit longer than we anticipated.

And the customer feedback has generally been quite good.

So we don't have anything to announce yet in terms of other markets, but it's certainly something we'll take a look at.

Great.

And then second, what percentage of the bookings are currently made through marriott.com, spg.com and the apps combined when you roll it all up?

And as you move to one platform, are you expecting that, that percent, which I think is growing each year, to accelerate in growth?

When you look at -- if you add property to it...

It's 3/4.

It's about 3/4 of our business is coming through our channels.

Maybe 70% would be our channels.

So that would be calling the hotel, calling our call centers and our digital platforms, whether they'd be the dotcom sites or the apps.

I think what we've seen over the last number of years is that the digital channels have grown substantially.

The voice channels, including our own voice channels, have declined.

We're likely to see those trends continue, and we do think for the reasons we discussed earlier about share of wallet, having all the hotels show up on one platform, that we should be able to drive enhanced growth in the digital platforms because everything will be there.

We don't have a forecast for you on where those numbers are going, though.

And your final question is from Stuart Gordon with Berenberg.

Net unit growth, obviously, there's 2 sides of the equation.

And everybody's been focused on the deletions.

But I was wondering on the growth side, you've materially increased both the number and proportion of rooms and construction.

So is there an opportunity for you to do better than the 285,000 to 300,000 gross addition number you gave at the Capital Markets Day?

Well, there's a few questions in there.

I would say, first, that we've been positively surprised by how strong the development pipeline is.

I think when we've -- this is a conversation we've been having for a few years, but we would have thought that U.S. signings and approvals probably would not be as strong as they are today.

We thought that 2016 was sort of the peak organic growth and it probably still is the peak organic growth.

I don't think necessarily, we're going to see the numbers move up from there.

But our partners in the United States continue to want to do more deals with us, and that's been really quite good.

The world, obviously, is a big place.

Some markets continue to perform quite robustly.

Some are essentially moribund in the development space either for economic reasons or for geopolitical reasons.

I think if you look at all those averages and compare to what we shared at the analyst conference, we would say that deletions might be just a little bit higher than we had before, but we'd have to go back and look at those 3 years.

And openings have shifted back by a quarter or 2 and that's because of length of the construction cycle.

And I think that latter factor is more likely to cause us to miss that 285,000- to 300,000-room number we used 1.5 years ago?

When was that conference?

It was March of '17.

March of '17.

More likely to miss it than we are to exceed it.

Okay.

And just a follow-up.

You've given some color before on the number of credit card customers you've got.

Could you give us an update on how that has improved with the better terms and also the cross-selling opportunities that you have as a larger group?

Well, it is -- it's still new in the process.

JPMorgan Chase has rolled out the new Visa card and the customer response has been great.

They're in market.

Amex will not roll out its new product until later this month, I think roughly the end of August.

I don't remember precisely the date.

And so we don't have the customer reaction there yet.

But I -- all things considered, we feel quite good about the strength of the program and the likelihood that it's going to grow in the future.

All right.

Well, thank you very much.

Thank you, everybody, for your time and attention on the call.

We appreciate your interest in Marriott and look forward to welcoming you into our hotels wherever your travel takes you.

Thank you.

Ladies and gentlemen, this does conclude today's conference call.

You may now disconnect.