PEB - NYSE

Greetings, and welcome to the Pebblebrook Hotel Trust' Third Quarter Earnings Conference Call. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Raymond Martz, Co-President and Chief Financial Officer. Thank you, sir. You may begin.

All right. Thank you, Christine, and good morning, everyone. Welcome to our third quarter 2025 earnings call. Joining me today is Jon Bortz, our Chairman and Chief Executive Officer; and Tom Fisher, our Co-President and Chief Investment Officer. But before we start, I'd like to remind everyone that our remarks are effective as of today, November 6, 2025. Our comments may include forward-looking statements that are subject to various risks and uncertainties. Please refer to our SEC filings for a detailed discussion of these risk factors and visit our website for reconciliations of any non-GAAP financial measures mentioned today. Now let's jump into the quarter. We're pleased to report that our third quarter performance was in line with our outlook in a challenging quarter shaped by heightened geopolitical and macroeconomic uncertainty as well as an unfavorable holiday calendar shift, we again delivered solid operating results and industry-leading cost controls. This execution sets us up well for 2026, given the robust convention and major event calendars across our markets. Same-property hotel EBITDA totaled $105.4 million, in line with our midpoint, while adjusted EBITDA came in at $99.2 million, exceeding our midpoint by $2.2 million. Adjusted FFO per share was $0.51, $0.03 above our midpoint. Together, these results reflect the resilience of our operating model, our relentless focus on driving operating efficiencies and our disciplined cost management. On the ground, performance was led by our properties in San Francisco and Chicago, alongside strong contributions from several of our recently redeveloped resorts, including Newport Harbor Island Resort and Jekyll Island Club Resort. Turning to portfolio trends. Same-property occupancy increased nearly 190 basis points, while ADR declined 5.4%, resulting in a 3.1% decline in RevPAR and a 1.5% drop in same-property total RevPAR. If you exclude Los Angeles and Washington, D.C., our 2 most challenged markets in the quarter, total RevPAR actually was up 0.6%. The decline in ADR was primarily driven by competitive pricing in D.C. and L.A., stemming from disruptions related to the ICE activity and the National Guard deployments. We also saw more demand coming through lower-priced booking channels, offsetting softer group attendance and government travel. Even so, occupancy increased in 6 of our 7 urban markets and across nearly all of our resorts. San Francisco was once again the standout. RevPAR rose 8.3% in Q3 on a 690 basis point jump in occupancy, driving EBITDA higher by 10.9%. Growth was broad-based with increases fueled by an active convention calendar and a continued recovery in both business travel and leisure demand. Results would have been even stronger but for the massive Dreamforce citywide convention shifting into October from September of the previous year. Importantly, positive momentum continues in San Francisco with a very strong fourth quarter well underway. No doubt, San Francisco has gone from a laggard to a leader led by the AI revolution, which is headquartered in the city and by San Francisco's tremendous progress in becoming a cleaner, safer and more vibrant city. Chicago also posted another solid quarter with RevPAR increasing 2.3% on healthy leisure events such as concerts and sports, improving weekday corporate travel and stronger weekend leisure. These positive results were achieved despite Chicago facing an extremely difficult comp to last year when the city hosted the Democratic National Convention in August. Both San Francisco and Chicago continue to pace well through year-end and into 2026, which reflects one of the many reasons we're more constructive on next year. Our resort portfolio also remained resilient with total RevPAR increasing by 0.7%, led by exceptional growth at Newport Harbor Island Resort, where RevPAR jumped 29% and total RevPAR surged an impressive 35.9% versus its pre-renovation performance in 2023. Jekyll Island Club Resort generated an 8% RevPAR increase with total RevPAR growing over 11%, while Estancia La Jolla's RevPAR rose 5.7%. These properties illustrate the power of our redevelopment program, which is driving market share gains and growing profitability as these properties climb towards stabilization. Across our urban markets, performance was more mixed. Urban total RevPAR declined 2.7% as strength in San Francisco and Chicago was offset by ongoing weakness in Los Angeles and Washington, D.C. and the lighter year-over-year convention calendars in Boston and San Diego. Washington, D.C. was our softest market, with RevPAR down 16.4% due to reduced government and government-related travel demand and lower tourism activity. We expect these challenges to persist throughout much of Q4 given the federal government shutdown. However, the setup in D.C. should improve significantly in 2026 with more normalized federal travel, a favorable convention calendar and numerous America 250 events. In Los Angeles, RevPAR declined 10.4%, driven entirely by rain. Greater price competition emerged from the negative impact of the devastating fires earlier in the year and the pressure did not decline in Q3 as the ICE rates and National Guard deployments created a perception of disruption and safety concerns, driving continued rate pressure. Conditions are stabilizing as the political environment cools and entertainment production gradually improves. So we expect L.A. to only be a minor headwind in the fourth quarter. Boston and San Diego experienced year-over-year declines attributed to lighter convention calendars in the city as well as softer group attendance. San Diego has also been negatively impacted all year by the significant cutback in federal government travel. That said, both markets continue to exhibit steady underlying trends in leisure and business travel. On a monthly basis, July same-property total RevPAR decreased 1.1%. August was essentially flat and September fell 3.3% with the midweek timing of the Jewish holidays being a major headwind for September as expected. On the revenue side, same-property out-of-room revenues grew 1.7%, supported by stronger event space utilization, elevated food and beverage performance and the benefit of upgraded amenities across our redeveloped properties. Transient demand strengthened by 3.8% in Q3 as the booking window remains shorter, aided by growth in our wholesale and consortia channels. Group occupancy declined by 2%, primarily due to lighter-than-expected attendance at health care, education and government attended or related events. This trend is consistent with the national data STR has been publishing, which we also highlighted last quarter. On the operating expense side, execution remained excellent. Same-property hotel expenses before fixed costs rose just 0.4% year-over-year. And on a per occupied room basis, expenses declined about 2%. That's another quarter of exceptional operating discipline by our hotel teams and asset managers, creating efficiencies and lowering operating costs. Turning to LaPlaya in Naples, Florida. Our weather resiliency improvements are just a week or 2 away from being substantially complete. We expect LaPlaya to generate approximately $36.6 million in adjusted EBITDA this year, including both hotel EBITDA and BI income. This compares to $42.8 million in 2024, which benefited from elevated BI collections following Hurricane Ian. On the capital side, we invested $14.2 million in the quarter and remain on track to invest $65 million to $75 million this year, reflecting a return to a normalized capital investment pace following our now completed multiyear redevelopment program. This lower run rate supports higher discretionary free cash flow and gives us more balance sheet flexibility. We also entered into an agreement to sell one of our hotels for $72 million with a buyer having provided a nonrefundable deposit under the contract. Consistent with the purchase agreement, we can't disclose a specific hotel or buyer at this time. The property has been classified as held for sale, and we expect the transaction to close in the fourth quarter, subject to customary closing conditions. That said, there is no assurance that the sale will be completed on these terms or the time. The potential disposition is not reflected in our fourth quarter or full year outlook. Shifting to our balance sheet. We remain extremely pleased with the successful $400 million offering of 1.625% convertible notes we completed in September. We used these proceeds to retire $400 million of our 1.75% convertible notes due 2026 at a 2% discount to par, leaving a very manageable $350 million outstanding. We also concurrently repurchased $50 million worth of common shares during the quarter at a significant discount to NAV, which is accretive to FFO and NAV per share. We ended Q3 with $232 million of cash, and we expect to generate over $100 million in free cash flow by the end of 2026. Our plan is straightforward: use cash on hand and free cash flow to take out the remaining convertible notes maturing in December 2026. All told, it was another quarter of disciplined execution amid a choppy and uncertain demand backdrop. And with that, I'll hand it over to Jon to provide more details on the third quarter, our outlook for Q4 and a look ahead to 2026. Jon?

Thanks, Ray. When we look at the industry's performance, the third quarter looked a lot like the second, only a bit softer. Demand was slightly down year-over-year, and that caused renewed pricing competition, which led to a lack of ADR growth. Group demand was most pressured. It was lower in all 3 months due to reduced government travel, weaker international participation at conventions and conferences and some increasing attrition. Transient demand, including leisure, held up better. It remained positive versus last year. That mix favored weekends over weekdays for the broader industry and for Pebblebrook. In terms of industry performance by price point or scale, there remains a sharp divide between the upper and lower ends of the market. Premium hotels and resorts continue to perform better, while the bottom half is seeing much more weakness as cost-conscious consumers pull back on their discretionary spending. In Q3, we faced the same fundamental challenges as the industry, but the localized disruptions in L.A. and Washington, D.C. drove our third quarter performance below the industry average. To put that disruptive impact into perspective, L.A. and D.C. represented roughly $7 million of the $7.9 million year-over-year decline in same-property hotel EBITDA. Throughout our portfolio, we continue to see a recovery in business transient travel. Like the industry, group room nights and group revenues were slightly negative in the quarter versus last year, while business and leisure transient demand continued to improve. Due to the resiliency of leisure demand, weekend occupancies were up all across our portfolio, urban and resort, demonstrating the continued appeal of our high-quality properties, especially for leisure and social group customers. Weekday occupancy also grew due to the continuing recovery in business transient travel and our team's focus on replacing group and government shortfalls and rebuilding overall occupancies through discounted wholesale and consortia channels. I'd also like to briefly highlight the performance at our redeveloped properties because it's a key part of our improved performance in '25, and it should provide a similar boost in 2026. We praised the terrific performance of Newport Harbor Island Resort last quarter, and it deserves that praise again this quarter. In Q3, Newport led the way in our portfolio, delivering $11.8 million of EBITDA in its most important seasonal quarter, up $2.9 million year-over-year on a 21.6% total revenue increase and strong flow-through. That's exactly the ramp we expected from the comprehensive $50 million transformation completed last spring. That's a higher quality overall resort experience with more compelling venues, delivering increased event capacity and a richer food and beverage mix, all together driving higher ADRs and higher out-of-room guest spend. For the full year, we now expect Newport to generate almost $17 million of EBITDA, ahead of the $13.6 million at acquisition and much higher than our forecast just 90 days ago. We're very excited about Newport's future. Hats off to the resorts operating team. And 2025 is just our first full year of post redevelopment operations. So we believe the resort is well positioned to generate even stronger performance over the next few years as it continues its ramp and it benefits from increased exposure to group and leisure demand. And Newport is just one example of the benefits of our strategic redevelopment program. Our comprehensively upgraded and transformed hotels and resorts across our portfolio are gaining share and growing cash flow with more runway ahead. This includes, among others, Estancia La Jolla, Chaminade Resort & Spa in Santa Cruz, Hotel

[Operator Instructions] Our first question comes from the line of Gregory Miller with Truist.

First off, very detailed and helpful commentary on 2026. I'd like to ask about San Francisco lodging performance given some encouraging trends as of late. I could ask several questions about how you see the market today, but I'll start with a few items in particular. Looking at CoStar data, there has been some encouraging room rate growth, especially during major convention citywides [indiscernible] and obviously, there are many potential citywide sellout days ahead in the next couple of quarters. I'm curious what you're seeing in terms of the level of confidence from hoteliers in the market to push room rates during high occupancy nights and any implications to your properties?

Yes, Greg. So thanks for the focus on San Francisco. Obviously, it's an important market for us, and it's not been a good market for many years. But this year, it's really cranking on all cylinders at this point. And I think your question about rate is really where the market is heading, which is -- and there's a big opportunity that I think has begun to already be capitalized on to push rate, not just over the citywide dates where compression clearly occurs, but much more so on weekdays and certain weekends when there are major events in the city. But as it relates to weekdays, I think what we're seeing is regular Monday, Tuesday, Wednesday nights without a citywide are still selling out. And that increase -- very significant increase in demand is allowing our teams to push rate and do it with increasing confidence. And I think as that permeates through the market, I think we'll continue to see that. I mean we have a lot of rate growth to recover from 2019. And I think we've started to see it, but I think that will be a much bigger part of where we go next year in addition to obviously being able to take advantage of both Super Bowl, which, of course, will drive significantly higher rates as well as World Cup, which will not only drive rates, but even more importantly, drive significant occupancy growth in the market.

And Greg, also just about your question of momentum building in '26. The convention center, the room nights on the books for '26, it's made a lot of really good progress since the start of the year, thanks to SF Travel is doing a great job with the new leadership there. Just to give you an example, they booked over 100,000 room nights or 100,000 room nights have gone definite for '26 since the start of the year. So that's about a 20% increase. So we started the year with '26 convention demand looking like it would be down in '25. It's actually now up in a pretty short amount of time. So it shows the positive momentum there. And the number of sellout nights we had, what, 44 days in '25 that we had over -- redeemed as kind of sellout nights, that's projected to increase in '26. So to Jon's point, we have more opportunities there for hull some great compression opportunities.

And I think one of the things that's kind of unique about San Francisco, obviously, you've got a big factor in that AI and much of technology and biomedical is headquartered there. But a lot of the conventions and citywides that occur in San Francisco are corporate led. And as a result of that, corporate tends to book much more short term than the bigger associations do. So bringing in that major corporate-sponsored events like Microsoft Ignite, which is occurring here in November, where they just canceled out of 2 other markets for '26 and '27 and have signed up to be here in San Francisco. So we're really encouraged about the positive momentum there.

That's all very helpful. Maybe switch gears on a different demand segment in San Francisco. We get a lot of investor questions that relate to transient corporate and specifically how the AI industry is impacting demand at this point or maybe the next couple of months. I'm curious if you could provide a little more context in terms of what you're seeing as of late and perhaps your expectations for the fourth quarter.

Sure. Well, I mean, what we've been seeing and increasingly over the course of the year are more and more companies that we've never heard of. And they're booking not just transient business, they're booking in-house group, they're booking, recruiting people coming in, looking for jobs. They're booking training in our hotels in the marketplace. And some of these have grown to the point where they have their own conferences like Snowflake, which I guess, didn't exist, I don't know, at least certainly people didn't know about 3 or 4 years ago. So it's definitely having an impact on those Tuesday -- those Monday, Tuesday and Wednesdays that I talked about within the market. And so, it's extremely encouraging. They're also coming in. They're taking a lot of office space. Obviously, people have read the stories about these companies growing and they're taking space quickly after they take space because they're growing so fast. The other thing we've seen is, a significant number of IPOs of companies based in San Francisco over the last 6 months. And again, that's capital flowing into the industry that's being used for growth, and that growth involves to a great extent, high-caliber people being hired.

Our next question comes from the line of Cooper Clark with Wells Fargo.

You continue to make really strong progress on the expense side. Could you provide color on how much of that is the result of reduced headcount? And is it fair to assume we see labor costs moderate into '26?

Yes. I mean I think -- I mean, I've been in the business since the early '90s. And through every cycle, we reduced the headcount of our hotels. Our people become more efficient, more productive, and that's continuing. There's additional tools that people can use. We're better at scheduling through utilization of these tools. We're making fewer mistakes. When it comes to scheduling, we're using third-party services less as a result of more efficient scheduling as examples. So we think that will continue. That's our relentless focus. And yes, I think the wage side because of the front-end loading of a lot of these city labor contracts, we think that next year, the wage rate growth will be less than it was this year, albeit a little of that will be offset by probably health care costs that are going to be -- that are going to grow at a higher cost -- a higher rate than they did this year. But overall, it should be lower in terms of the growth rate of those wages and benefits combined. But our focus is on everything that we do in our hotels. It's on how are we more efficient utilizing energy. There's new tools. There's better ways to operate our hotels to use less power, to use less water. It's good for the planet. It's good for the bottom line financially at the end of the day. It's what our customers want. It's more consistent with their values. It's focused on insurance, how do we lower insurance? How do we reduce accidents? Some of that is operating best practices. Some of that involves physical improvements and changes to our properties. How do we increase resiliency from weather? So we have less downtime and less damage. It's infrastructure improvements, roofing, window ceiling, new windows, sealants. It's all sorts of things through the portfolio. It's how do we reduce credit card commissions, how do we encourage people to pay by ACH? All of that is a focus. It's really every line item that we have on our income statements. And it's going to be a significant focus as far as the eye can see, particularly as we get new tools that are AI-enabled or boosted, robotics become a more important factor. And as the quality of that service becomes better, in fact, in many cases, likely better than what a human can provide ultimately. So certainly more knowledge individually. So we're very encouraged about where the opportunities are to lower costs as we move forward, improve the quality of work for our employees and deal with some of the shortages that are occurring and that we think will occur as a lot of our workforce ages and is not being replaced by others willing to do the same jobs.

Okay. That's really helpful. And then shifting to L.A. I appreciate some of the more positive commentary on L.A. into the fourth quarter. Could you just talk about how we should think about L.A. into 2026, given what should be softer comps, but some continued challenges on the demand and labor side of things? I guess, said differently, do you expect L.A. to continue to drag on results from a RevPAR and EBITDA perspective over the next 12 months relative to the rest of your portfolio?

Yes, Cooper, we -- well, we think actually it will be one of our better performers next year because of the easy comps, the recovery that we are seeing sort of renormalizing back in the market. And then the discussion and data that we're getting related to the future ramping of TV and movie production in the state based upon the approvals of those that are going before the film commission that's providing these grants. From what we see, the big ramp-up of that, there's a small increase that we should see probably not here in the fourth quarter, but in the first quarter of next year, but then a bigger ramp in the second quarter because there's a 6-month requirement that once the application has been approved that they start production in the market. So the doubling -- more than doubling of the credits that the state is providing for production in California is really kicking in by the middle of next year. So look, presuming we don't have any major events, we don't have political issues going on that disrupt the perception of safety or quality of life or the beauty of the visit. We think L.A. will be a big tailwind for us in the portfolio next year.

Our next question comes from the line of Smedes Rose with Citi.

I appreciate all your detail talking about next year. But I wanted to just ask you a little bit about maybe -- I mean you have an asset held for sale. Just kind of what are you seeing overall in the transaction market in terms of just, I guess, overall pricing? And where are you, I guess, going forward in terms of trying to execute on potential asset sales, assuming there's kind of a constructive transaction market?

Smedes, it's Tom. Thanks for the question. So I think just as a general backdrop, the transaction market has kind of been gyrating between risk-on and risk-off all year. The one constant throughout the year, however, has been the debt markets. The debt markets have been improving. They're getting more competitive. There's more availability, there's better pricing. And in some instances, it's actually becoming an alternative to a sale for many sellers from that perspective. Because I think on the equity side, again, it's kind of more risk-on, risk-off. I think would be government shutdown, would be kind of flat to negative operating performance that you see in the weekly and monthly stars. I think star reports, we're just kind of at a pause right now until there's a little more macro clarity. But what I would say to you is, over the course of the last 60, 90 days, we've seen, I think, a real pent-up demand by investors. You've seen some larger transactions take place. You've seen a return of some of the bigger private equity names. You've seen some of the owner operators. So I think there -- my sense here is that, there's just -- they're all just waiting for that catalyst. And I think if you listen to what Jon had indicated in our call about 2026, I think once things turn and there's better visibility, I think there's going to be a lot of pent-up demand for transactions moving forward. I think the risk-off situation right now is nobody really wants negative leverage and they're focused on smaller deals and those will continue. But until there's some clarity, I think we're going to be a little bit of a pause here.

And I think from a strategy perspective, our focus continues to be to sell assets and use that capital to take advantage of the public-private arbitrage opportunity to buy our stock back, pay down debt, remain leverage neutral or slightly reduce leverage over time. But I think from the disposition perspective, there have been new entrants into the market. There's certainly a lot more high net worth individuals out there looking at lodging where they might not have previously because I think folks see the potential upside opportunity and the ability to take advantage of what are historically pretty low per key values, particularly as it relates to what has been continuously increasing, which is the replacement cost of hotels. So I think we're encouraged, but I think what we need, and we've continued to need and we've talked about this before is, we need operations to turn positive. It's hard for a buyer to buy into a market that's declining. They need to see things go up. Everybody is well aware of the long runway of limited supply growth, which will allow for both occupancy and rate to grow arguably faster than inflation, which is what it's done historically. But it's got to turn. So I think that's where we're going. And hopefully, we have fewer of these macro disruptions next year.

That's helpful. And then I was just wondering, just -- you mentioned that you continue to see, which we hear generally across the states, a real discrepancy between higher-end leisure customers and then lower end. And just within your portfolio, what are a couple of examples, I guess, of sort of higher-end resorts where you saw solid maybe RevPAR gains year-over-year and maybe a couple where they were weaker, just given the composition of the customers that go to those properties?

Smedes, well, there's a couple of other moving parts. I mean you take Newport Harbor as an example, which caters to people who are from Boston and New York. So we have a very strong leisure component there and also very strong corporate demand. Our demand has been tremendous there. As we cited during the call, the RevPAR up double-digit, 30-plus when you look at those levels over year-over-year. So those are examples where we're also benefiting from the redevelopment side there, but that's where we continue to have less price sensitivity. When you get to some of the markets, maybe in some markets, say, like in South Florida for Key West, there's a little more pricing sensitivity. But we've adjusted our revenue management strategies accordingly. So we actually did okay there. But like I think we're opening up channels to look at other sort of customer bases and doing our best there. But I think overall, the quality level of our hotels are so higher. We're a little more insulated versus if you start going down the quality spectrum, and you're seeing in STR numbers much more than our resorts.

And I think a couple of other examples would be LaPlaya in Naples and Inn on Fifth in Naples, both luxury properties, a fair bit of insensitivity to pricing by our customers down there. L'Auberge Del Mar in Del Mar out in California would be another example of a property at much higher average rates, relatively small property where the customers, again, are relatively insensitive to pricing, but very sensitive to us providing them good service. So those would be a couple of other examples.

Our next question comes from the line of Duane Pfennigwerth with Evercore ISI.

Jon, on government shutdown impacts, given your time in D.C., any perspective on how quickly this activity can spool back up? Because if I think about this year, we already absorbed the DOGE impacts earlier in the year. I assume you had won some of that back, but maybe not all of that back, and now we have the shutdown. So I don't know if you covered it in your extensive 2026 event navigation, which was very helpful. But have you sized the impact from the government sector this year in totality? And how big of a tailwind could it be next year?

Yes. It's funny. These things are a little harder to estimate. I mean we can look at where government is, and it's probably been down about 1/3 to 40% of what it was the year before. That probably represents, in total, about 1 point to 2 points of total demand for our portfolio and probably fairly similar for the overall industry, although probably more heavily weighted at the low to middle end, generally speaking. The shutdown is a different matter because it's impacting more -- way more than just government travel. It's impacting people who have discretionary travel, which is what most travel is at the end of the day. So it's more about people who get anxious about flying. It's more about people who don't want to put up or take the risk of cancellations or big delays that they're concerned about and that the media will be keen to report about. So it's a hard thing to measure, but it's material as is the continuing imbalance of international -- domestic outbound and international inbound. I mean you've got one that's at 120% of 2019 levels, and you have inbound in the 80s of 2019 levels. That could reverse, but it's going to take it sort of creating a different perception on the part of our government's desire to welcome people into the country. And frankly, we're not seeing that yet, although we hope to see that certainly for World Cup next year, which is something that's really important to the President and the administration.

And then just relatedly, I ask you the same question I asked was, given that the assumption that we have a no storm fall and we're not like rebuilding this fall on the Gulf Coast, what does that allow you to do and thinking more about like 2026?

So I think the first thing it allows us to do is sell a property that's not been damaged. One of the things we were hearing from clients in Naples, as an example, is, gosh, you've had all these storms in the last few years. We've had to move our meetings to other markets or other properties. Is this going to ever end? Or is this the new pattern? And having a year where there's no impact, I think, is helpful from a sales perspective. Certainly easier to sell when you don't have that. It's easier to sell when you have a property that's primed and in great condition, which is not what we had last year in terms of selling for this year. So I think it bodes well for, again, LaPlaya to ramp from $25 million of EBITDA this year to maybe closer to $30 million next year on its way to hopefully getting back to that $35 million or $36 million level of where it was heading in 2022.

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

Jon, you mentioned attrition or increased attrition during the quarter. Can you dig in there a little bit more? What markets -- what segments did you see that? And presumably, that's continued into the fourth quarter? Just any added color on short-term booking trends would be helpful.

Yes. I mean it varied through the portfolio in markets, but it was much more visibly related to government and government-related, but you don't always know what's government-related until the customer declares it. And in a lot of cases, it involves education. A lot of these conferences are supported by grants or the departments are supported by grants, which they're not getting, they're frozen or they don't expect to get, as you know, with the disruption going on with the universities. So a lot of it is related to those sectors. We're not really seeing it in technology. We're not really seeing it in medical, biomedical. We do see it some in other conventions where the attendance is lower and it relates to associations where the people are paying their own way. And you get some of that 3%, 4%, 5% falloff in attendance that results from that. So it's not something that -- I mean, interestingly, I think still on a year-over-year basis, our attrition payments were less this year than they were last year for Q3. So it's certainly not at a high level yet, but it's more clearly impacting those groups that are somehow related to government.

Yes. So, Michael, we're not highly concerned with the attrition cancellation. It's not heading in a really bad direction, but it's certainly something we do monitor because it does go quarter-to-quarter because that's usually maybe early canary in the coal mine, so to speak, if companies are feeling differently about their spending. So nothing materially that we're concerned about, but we continue to monitor it.

Our next question comes from the line of Jay Kornreich with Cantor Fitzgerald.

I just wanted to follow-up on the leisure transient side of the portfolio. The recently renovated resorts have all been performing quite well. But just curious, how would you characterize kind of the overall leisure customer and its price sensitivity these days? And as you look out towards next year, on the same-store part of the portfolio, do you feel like there's more upside from that leisure customer improving or more from the urban side of the portfolio?

Well, interestingly, I mean, the leisure customer impacts our urban properties to a meaningful extent. I mean our cities are, in many cases, very heavily tourist destinations. And when I think about -- when we think about next year, I would say a couple of things. First of all, some of the few cities like D.C., as an example, and L.A., leisure has been meaningfully impacted this year. We should see a recovery next year. I mean when the government shuts down, the Smithsonians closed. Museums are closed here in town. There's not a lot to see here if you were a leisure guest. If you're a group coming from a high school or a middle school somewhere coming to spend their 3, 4, 5 days in D.C., they're not coming right now because there's nothing to do. So that's definitely something that should be a tailwind unless government is going to be shut down next year. And so, I think from a leisure perspective, as the economy improves, and leisure customers feel better about their jobs, I think leisure will continue to improve. I mean I think it's lost on people that if you -- we were trying to highlight this. If you look at the weekends in the third quarter, they were up year-over-year in occupancy and demand. So the leisure customer has been resilient, but they have become more price sensitive. And as you work your way down the price spectrum, they get more and more -- the customers at those properties are more and more price sensitive. And you're seeing it in the differing rate declines that STR reports every week and every month.

And Jay, I think where we also counter our portfolio is a little bit different is all the redevelopment that we've completed over the last couple of years, that's paying huge dividends. So that's fighting through maybe some of the weaknesses that could be some of these consumers. If you look at our projects that we completed in '23 and '24, we've gained over 700 basis points of penetration over -- year-over-year. So that's a really a good direction there, which helps us counter maybe some weaknesses you may have in individual consumers.

Our next question comes from the line of Aryeh Klein with BMO Capital Markets.

Jon, I appreciate all the color on the disconnect between demand and GDP growth. And I was hoping you can touch more on why you think that disconnect has happened? And what gives you the confidence that, that resolved? And I guess alongside that, do you think that the growth in AI investments where building data centers doesn't provide all that much benefit to lodging demand has distorted the relationship between the 2 and just that maybe the underbelly of the economy isn't all that healthy that can continue into next year?

Sure. Well, I think the disconnect has a lot to do with some of these major policy disruptions and it has to do with sort of the normalization following the pandemic. I mean we had a big demand recovery. We had customers back in '22 as an example, that had no pricing sensitivity whatsoever. And you remember those comments we made about people buying up for suites, and that was the first product that went in our property and nobody really cared what anything cost because we couldn't even meet all the demand that was there. We didn't have the staff to service it. That had to normalize. And I think the other thing that's had a negative impact on that correlation is this international domestic imbalance, which is really large. And keep in mind, an individual who goes abroad or comes here is usually spending -- I mean, I think here, it's -- the average is somewhere between 10 days to 2 weeks abroad. I think it's a little more like 7 to 10 days of the outbound. But that differential has gotten really, really wide post pandemic, and it's not recovered. And so, I think that's a big part of it, too, Aryeh. I don't think it has anything to do with data centers and the GDP really isn't as good because you have to -- these things -- it's about direct and indirect, right? There's a lot of money being invested. Where is it ultimately going? Ultimately, it's going to go to people and businesses and corporate profits and wages and benefits and bonuses. And so, ultimately, it gets around to enhance the economy. And that -- it doesn't really matter all that much what it gets invested into unless it's -- I don't know, it would have to be something that went out of business really rapidly. But -- so I do think all that capital, I mean -- and it's not just data centers. You've got a lot of manufacturing reshoring. You have to have materials for that. You've got to ship that stuff. You've got to put it into place. You have to hire a lot of workers to do it. You have to rent a lot of equipment to build those things. You've got to build a lot of new electricity and buy all the equipment. I mean it goes on and on and on. And you still have the vast majority of the CHIPS money that hasn't gone out the door yet. You have a lot of the infrastructure bill money. I mean, you can just go online and ask how much of that money has gone out the door. The majority of it still has not gone out the door for infrastructure, but it's been authorized. So a lot of these disruptions, hopefully, international begins to normalize over time. It's not in our thoughts for '26. You noticed we didn't mention of that reversing. I mean the dollar, when it retreated in the first half of the year has recovered a lot of that retreat. That's not good for international inbound and it hurts outbound. So I think that's in the pocket. It's something that will ultimately normalize, but I wouldn't count on it for next year unless we see a reversal in the dollar. And some of this other rhetoric that's caused people to go elsewhere.

Maybe just a quick follow-up. Just for the World Cup for next year, any early thoughts on the magnitude or the potential tailwind that, that can add to RevPAR for next year?

Yes. I mean I think some of the brands or prognosticators have suggested it's maybe 30 basis points or 40 basis points of improvement in RevPAR next year. I tend to think it's probably a little more than that for our portfolio given the number of matches that we have. The challenge is, I mean, we've gone -- we've done a lot of research on it with our teams. And if you go to the last World Cup, 70% of the business was booked within 30 days of the matches. What gets booked ahead of time, and we have started to see some of it is some of the group that is regardless of what team is playing and where they're playing. So we would expect it to be very last minute. We haven't announced the teams yet. There are some, obviously, that are -- that have already qualified, but still the majority have not. Not only have they not determined all the teams, but they haven't said where any of the teams are going to be playing. So all of that just means it's going to be pretty short term. But the other positive is, there's 50% more teams this year than there were last year, I mean, in the last World Cup. We've gone from 32 to 48 teams. That's also a big positive for the overall impact. So I think it will be -- I think a lot of it's going to happen late first quarter and second quarter and really the final 30 days before the matches in both June and July. And that just means everybody is going to hold out and keep things frozen until then.

Our next question comes from the line of Chris Darling with Green Street.

Going back to San Francisco, we've obviously talked about how the market has plenty of momentum behind it. I think importantly, it seems like investor sentiment has really changed for the positive as well. With that in mind, Jon, curious where your head is at strategically in regards to your remaining exposure there. Do you think now might be the time to consider divesting some of your remaining assets? Or does it make more sense in your mind to sort of ride the recovery wave out over the next couple of years?

Well, I think from a general perspective, all of our hotels are available for a buyer, particularly strategic buyers because of the flexibility of our properties, all of our properties in San Francisco can be available without management and brand. So ultimately, there's a lot of flexibility there. I think where we are is, it will depend on pricing. I mean there's going to be a really strong growth. We believe in that strong growth. I mean we think we have assets that we think will double or triple their yields over the next 3 years in that market. And we believe you could easily see double-digit RevPAR growth for the next 3 to 5 years there, given the recovery that's needed in rates, the momentum that's going on with the underlying industries and growing confidence as occupancy gets rebuilt here in the market. So we've got great political leadership there. At this point, I think it will just depend upon pricing, whether we would sell in that market or not. We would just have to take into account what we believe the growth levels are going to be.

Mr. Bortz, we have no further questions at this time. I'd like to turn the floor back to you for closing comments.

Thanks, Christine. Thanks, everybody, for participating. Sorry, we ran long. We had a lot of thoughts we wanted to convey. We look forward to talking to you in February next year, but I'm sure we'll speak to many of you at NAREIT in Dallas next month. Thank you very much.