Summit Hotel Properties, Inc. (INN) Q3 2022 Earnings Report, Transcript and Summary

Good day, and thank you for standing by. Welcome to the Summit Hotel Properties Q3, 2022 Conference Call [Operator Instructions]. Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Adam Wudel, Senior Vice President of Finance, Capital Markets and Treasurer. Please go ahead.

Thank you, and good morning. I am joined today by Summit Hotel Properties' President and Chief Executive Officer, Jon Stanner; and Executive Vice President and CFO, Trey Conkling. Please note that many of our comments today are considered forward-looking statements as defined by federal securities laws. These statements are subject to risks and uncertainties, both known and unknown, as described in our SEC filings. Forward-looking statements that we make today are effective only as of today, November 3rd, 2022, and we undertake no duty to update them later. You can find copies of our SEC filings and earnings release, which contain reconciliations to non-GAAP financial measures referenced on this call on our Web site at www.shpreit.com. Please welcome Summit Hotel Properties' President and CEO, Jon Stanner.

Thanks, Adam, and thank you all for joining us today for our third quarter 2022 earnings conference call. We were encouraged by our third quarter results, which reflect the ongoing improvement in our operating fundamentals, as same store RevPAR recapture was 95% of 2019 levels for the quarter, a 120 basis point increase from the second quarter and a new quarterly high for our portfolio. The quarter was highlighted by a particularly strong September when both same store and comparable 2019 portfolio RevPAR recapture reached 98%. This represented our best operating performance of any month since the onset of the pandemic, as business transient, midweek and urban demand improved meaningfully post Labor Day and are increasingly driving growth in our business. Top line growth continues to be driven by strong pricing power, as average daily rates are now consistently running higher than 2019 levels across nearly all segments of our business. For the third quarter, same store and comparable portfolio ADRs were approximately 4% higher than 2019, both of which represented an approximately 200 basis point improvement from the second quarter. While ongoing labor challenges persist in our industry and rising utility costs, combined with a particularly hot summer across the Sunbelt, put pressure on operating margins. Hotel EBITDA recapture reached 92% in our same store portfolio, an improvement from the second quarter despite seasonally driven lower nominal rates and RevPAR quarter-over-quarter. Once again, September's results highlighted the quarter, as hotel EBITDA was 6% above 2019 in our same store portfolio. The first month hotel EBITDA was ahead of 2019. Same store property level EBITDA margins exceeded 2019 levels by nearly 300 basis points in September. We continue to see evidence that the recovery in our business is increasingly being driven by business transient and group demand, which are supplementing what has been and continues to be a robust recovery of leisure travel. RevPAR recapture in our urban portfolio was nearly 90% for the quarter and heavily business transient-driven markets such as Boston, Charlotte, Chicago and Pittsburgh, all achieved RevPARs that exceeded 2019 levels during the quarter. While still trailing 2019, rates in our negotiated segment were 95% of 2019 levels, a 300 basis point sequential increase from the second quarter. Average weekday rates during the third quarter fully recovered to 2019 levels for the first time, driven by outsized growth in our urban portfolio. Weekday ADR in our urban portfolio accelerated throughout the third quarter, finishing September a robust 9% above July and 17% higher than September of last year. Preliminary October RevPAR of $130 for our comparable 92-hotel portfolio is another new pandemic era high and represents a 96% recapture to October of 2019. Growth in October again reflects recovering business travel as occupancy reached 72% in our urban portfolio for the month and urban RevPAR increased nearly 14% over September. Midweek demand also continued to strengthen in October, as Tuesday and Wednesday RevPAR recapture improved sequentially by 300 basis points and 400 basis points, respectively, compared to the third quarter. October is historically the strongest month for our portfolio in the fourth quarter, as normal seasonal trends lead to lower November and December nominal RevPARs. However, 2019 RevPAR recapture rate for the combined November and December period are expected to be in line with October levels, implying fourth quarter recapture rates generally in line with what we achieved in the third quarter. Likewise, we expect hotel EBITDA recapture rates to also be generally in line with the third quarter. In our earnings release last night, we announced the acquisition of our first high-end glamping asset, a distinctive 11 unit property in Fredericksburg, Texas, the epicenter of the Texas Hill Country wine region. Evolving our real estate strategy, along with emerging guest preferences has always been a hallmark of our capital allocation strategy, and we believe the demand trends that have elevated glamping from a niche travel market to an institutional asset class are poised to continue their rapid acceleration, as robust and resilient leisure demand continues to favor unique and experiential accommodations. Glamping has been the fastest-growing accommodation segment in the United States, as revenues grew by nearly 9% on a compound annual basis from 2017 to 2021. Glamping demand is expected to grow by nearly 15% on a compound annual basis between 2022 and 2030 driven predominantly by younger millennial and Gen Z travelers, which are projected to make up more than 75% of the glamping market by the year 2030. We view glamping as highly complementary [too] and a natural extension of our core business of owning high-quality hotels with efficient operating models and will benefit from our key operating competencies around asset and revenue management, as well as design, renovation and construction expertise. Glamping properties feature a labor light operating model, typical -- typically only 1 or 2 on-site staff members per site, and properties generally have few, if any, costly ancillary services and amenities. The uniqueness and experiential nature of glamping drives strong pricing power and combined with the labor light operating model drives particularly compelling unit level economics. We are underwriting unlevered glamping returns to IRRs that are 500 basis points to 1,000 basis points higher than a traditional hotel investment, driven by a significantly higher margin profile and profitability per unit, which can be 4x to 5x greater than a typical hotel EBITDA per key. We've structured an exciting programmatic partnership with Onera, an experienced developer, owner and operator of glamping and short-term rental accommodations throughout North America, designed to be a growth pipeline for the Company. The partnership will develop high-end experiential glamping properties in targeted markets across the country with our portion of the development capital funded primarily through our mezzanine lending program. As we did very successfully with our recent acquisition of the AC Element Miami Brickell, our mezzanine loans will be structured to provide an attractive in-place yield during the development period with an option to acquire each property upon completion at a predetermined value within a 90/10 joint venture with Onera. We have an exclusive right of first refusal on the next 10 Onera-branded development opportunities and have several exciting projects expected to break ground within the next 12 months. Trey will discuss the specifics of the Onera Fredericksburg acquisition shortly. And with that, I'll turn the call over to our CFO, Trey Conkling.

Thanks, Jon, and good morning, everyone. From a segmentation perspective, Summit's 42 hotel urban portfolio continued to see robust growth in both weekday and weekend demand, resulting in urban RevPAR growth of 27% versus third quarter of 2021. From a weekday perspective, urban demand continued its strong recovery with occupancy increasing by 770 basis points and ADR increasing by 19% from the prior year period. This growth was driven primarily by increasing business travel and group demand. From a weekend perspective, urban performance also continued to accelerate, driven primarily by ADR growth of 12% from third quarter 2021. In addition, for our urban hotels with comparable 2019 data, third quarter weekend ADR and RevPAR surpassed 2019 levels by approximately 16% and 2%, respectively. For the four week period following the Labor Day weekend, the 92-hotel comparable portfolio experienced notable strength in weekday demand, in particular, Monday through Wednesday, which realized occupancy expansion of 510 basis points versus the first 2 months of the quarter. Business Travel served as the primary catalyst of post-Labor Day performance, as midweek recovery in our urban portfolio increased 270 basis points in occupancy and $17 an average rate versus July and August. Strength in our non-urban portfolio was driven heavily by our hotels in suburban and airport locations. Our non-urban portfolio experienced an increase of 112% in weekday RevPAR compared to the post Labor Day period in 2021. Pro forma hotel EBITDA for the third quarter was $61.1 million, a 24% increase from the third quarter of last year. While labor markets continue to be challenging, we have seen stabilization in wage growth, while other operating costs appear to be moderating. Our same store and comparable hotel EBITDA margins were 120 basis points and 170 basis points below 2019 levels, respectively, which is in line with the second quarter margin differential despite lower nominal RevPARs. However, September margins for both same store and comparable portfolios were 275 basis points and 240 basis points higher than 2019 levels, respectively, partially driven by property tax rebates. Third quarter adjusted EBITDA was $47.2 million, an increase of 39% from a year ago. Adjusted FFO in the third quarter was $30.9 million or $0.25 a share, an increase of $10.4 million from the third quarter of 2021. Adjusted FFO will continue to benefit from our recent hedging activity and the high fixed nature of our capital structure. From a capital expenditure perspective, during the third quarter, on a consolidated basis, we invested approximately $24 million in our portfolio, bringing our year-to-date total to $49 million. Third quarter spend was primarily driven by several transformative renovations within our portfolio, including the SpringHill Suites Nashville MetroCenter and Hilton Garden Inn Houston Energy Corridor, which completed renovations in the quarter, as well as significant ongoing renovations at the Staybridge Suites, Denver-Cherry Creek, Hyatt Place, Orlando Universal Studios and Residence Inn Portland Downtown. Additional CapEx dollars went to purchasing for future renovation projects and routine maintenance capital. For the full year 2022, we expect to spend $70 million to $80 million on a consolidated basis or $60 million to $70 million on a pro rata basis in capital expenditures. As Jon mentioned, we are enthusiastic about our new partnership with Onera Escapes. Our initial investment is the acquisition of the Onera Fredericksburg, an 11-unit high-end glamping property in Fredericksburg, Texas, which is a popular year-round destination in the Texas Hill Country located within a 90-minute drive of Austin and San Antonio and within a 5-hour drive of Dallas and Houston. The 11-unit property consists of 10 unique temperature-controlled unit types with a mix of hard and soft-sided accommodations. Each unit has a private kitchen, bathroom with shower and most have private patios and hot hubs. Onera opened the property in November of 2021, and it has ramped quickly, generating year-to-date RevPAR as of September of approximately $425. Preliminary October results indicate the property's strongest months since its opening, with occupancy of approximately 90% ADR of $675 and RevPAR of $610. The property is expected to generate EBITDA margins of over 55% in our first year of ownership with EBITDA per unit of approximately $80,000. This translates to an NOI yield of 15% to 17% on the joint venture's total purchase price of $5 million and a 5.7x EBITDA multiple. Upon stabilization, EBITDA margins and EBITDA per unit are expected to be approximately 60% and $90,000 per unit, respectively. The joint venture also acquired an adjacent 6.4 acre land parcel for a total cost of $770,000, on which we plan to develop an additional 15 to 20 units at an attractive basis with comparable unit level economics. While the initial investment is small relative to our enterprise value, we are confident in the partnership's ability to scale quickly. We have identified several near-term development projects, inclusive of the planned expansion of the Fredericksburg location. These future projects are expected to generate similar RevPARs, margins and profitability per unit, as the Onera Fredericksburg property, with targeted opening dates in the next 24 months. Additional detail on the Onera Fredericksburg acquisition and joint venture are included in the investor presentation filed with earnings. Since July 2021, we have acquired 31 hotels, excluding the Onera Fredericksburg. For those hotels opened for the full year 2022, we continue to forecast a blended EBITDA yield of approximately 7%. The NewcrestImage portfolio is beginning to benefit from Summit's sophisticated asset and revenue management strategies, as third quarter ADR recapture improved meaningfully for hotels with comparable 2019 data, achieving a 104% recapture to 2019 compared to 100% in the second quarter. RevPAR index for the NewcrestImage portfolio also improved significantly, increasing 700 basis points over the second quarter. As a reminder, nearly half of the 31 recently acquired hotels have opened within the past 4 years, implying considerable upside in many of these assets. Turning to the balance sheet. Our current overall liquidity position remains robust at more than $460 million, which positions us well for future growth. During the third quarter, the Company defeased 3 CMBS loans totaling $55 million that were scheduled to mature in 2023. The defeasance event extinguished all but one remaining 2023 debt maturity, released $20 million of restricted cash and is estimated to generate net interest savings of $1.3 million through the scheduled maturity date. Additionally, the Company intends to defease the remaining 2023 debt maturity, a $32 million CMBS loan maturing in August of 2023 during the fourth quarter, which will eliminate all remaining debt maturities until the fourth quarter of 2024 after consideration of extension options. This will result in only 10% of Summit's pro rata debt maturing between now and year-end 2024, thereby significantly limiting refinancing risk given current volatility in the debt capital markets. The defeasance will also unlock $7 million of restricted cash, generate net cash savings of approximately $300,000 and result in AFFO accretion over the next 6 months. From an interest rate risk management perspective, our balance sheet is well positioned, including an average pro rata interest rate of 4.4% and approximately [67%] of our current outstanding pro rata debt fixed after consideration of interest rate swaps. In addition to address the pending maturity of $200 million in notional swaps, we entered into 2 $100 million interest rate swap agreements that will fix 1 month SOFR and carry fixed rates of 2.6% and 2.56%. These new swaps will mature in January 2027 and January 2029. This extends the average duration of our swap portfolio from less than 2 years to over 4 years. The swaps will become effective in January of 2023 after the $200 million of existing interest rate swaps expire. The new swap transactions will result in the Company maintaining approximately 70% fixed rate debt. And when including the Series E, F and Z preferred equity within our capital structure, we are approximately 75% fixed. On October 28th, our Board of Directors declared a quarterly common dividend for the third quarter of 2022 of $0.04 per share or an annualized $0.16 per share. The current dividend represents a prudent AFFO payout ratio, leaving ample room for meaningful increases over time. Included in our press release last evening, we provided 2022 guidance on certain non-operational items, including cash corporate G&A, interest expense, preferred dividends and capital expenditures, both on a consolidated and pro rata basis. We expect the midpoint of consolidated cash corporate G&A to be $21.5 million; interest expense, excluding the amortization of deferred financing costs to be $60.5 million; Series E and Series F preferred dividends to be $15.9 million; Series Z preferred distributions to be $2.3 million; and pro rata capital expenditures to be $65 million. With that, we will open the call to your questions.

Thank you [Operator Instructions] Our first question comes from the line of Neil Malkin with Capital One.

Good morning, everyone. Good to be with you. First question, I guess, the Onera, I mean that looks phenomenal. Very interesting. I think that maybe a year ago or so, you talked about thinking about some sort of alternative allocations within the hospitality segment in line with changing customer trends and demand. So maybe can you just talk about how that came to fruition? How long you guys have been in contact? And then maybe if you could quantify as best you can, what's the total investable universe or opportunity looks like with Onera or this type of lodging alternative.

First of all, thanks for the sentiment, and thank you for the question. I think you alluded to it correctly. This is something we've been working on and thinking about for a long period of time. I think if you go back and look at kind of the hallmark of our capital allocation strategy, it's always been around evolving our own real estate portfolio around, where we see changing and emerging guest preferences. And we've been very, very intentional [about] really going back to the IPO. And we've talked over and over again about the number of assets that we've sold out of the IPO portfolio and how many assets we own today. And we've turned over almost the entire portfolio, mostly again, driven by making sure we own a high-quality hotel -- a portfolio of hotels that cater to changing guest preferences. We really started looking at continue to evolve the portfolio even pre-pandemic. The process with Onera started in earnest probably about a year ago. So this is something we spent a considerable amount of time, one, trying to find the right opportunities, the unit level economics that are the most compelling and then probably very importantly, finding the right partner. And we do think we've got a really high-quality partner with the folks at Onera. So we're incredibly excited to work with them. We have mentioned that we do have a pipeline out there. We haven't quantified it. We do have several projects that we think are closer and more likely to get traction even in 2023. We do have a ROFO on the first -- excuse me, on the next 10 Onera-branded projects. These projects are $20 million to $25 million type project generally depending on the site, they can be bigger or smaller. But I think that's generally the rough approximation of where most of the projects come out. So we've tried to -- we've obviously started small here. Again, we wanted to do something, where we could prove the investment thesis out relatively quickly, but also something that gives us scale to grow this over time, as the business proves itself out.

I mean I'm sure other people have questions about it. I'll let them ask that, but those economics look compelling, maybe you can just kind of trade all your hotels for some of these and then just get to 17% NOI yields, that would be great. Other one for me is on BT. Obviously, the segment that has the most uncertainty around it still. I believe you said -- you gave a data point on the BT recapture versus '19. Correct me if I'm wrong, I thought you said it was about 90%. But more importantly, can you just talk about the -- how you see that the BT segment in your portfolio kind of progressing through the latter part of this year and into '23? And then if you could specifically focus on how you see challenged markets like San Fran and Chicago kind of progressing or shaping up through next year?

Well, look, again, we've talked a lot about how we think kind of the next leg of growth in our business is more BT driven, and it's more urban and midweek driven. And we've seen that repeatedly in our results. And I think we've seen it particularly in September and even our results into October, where September, we got to a 98% recapture rate. We actually exceeded our 2019 levels on the bottom line in September. That momentum has very clearly continued into October, where we had our highest nominal RevPAR of any month since the onset of the pandemic and we -- recapture rate again, at about 95%. Again, more and more of that is being driven midweek. More and more of that is being driven in urban markets. In markets that we mentioned a few in our prepared remarks that are specifically driven by BT customers, Boston -- markets like Boston and Pittsburgh and Charlotte and Chicago. So we've seen very good traction in those markets. I think our expectation for that business to continue to grow. We saw a very V-shaped [type] recovery on the leisure demand side. It's been a much more gradual recovery clearly on the BT side, but a recovery that we do think is ongoing, that we do think, to the extent there has been an inflection point in that recovery of demand, it has been kind of in this post-Labor Day period, where we've seen, again, much stronger BT performance. From a market perspective, I think, look, certain markets, and I alluded to them, markets like Chicago, Boston, even some of the suburbs of these markets, the BT business is, is really, really strong and driving some of markets that have lagged the first 6 months of the year have really had a strong post-Labor Day period. I think San Francisco is probably the outlier there. I think we've still seen San Francisco, as a slower market to return. They did have their first big meaningful city-wide out there in September. I think attendance was good. I think there's good momentum in San Francisco. But I think that's clearly a market that has lagged the recovery of some of the other gateway cities.

I guess, would you just expect that to be the case until sort of some structural things in San Francisco abate or are fixed? And then does that portfolio allocation, does that kind of go into it at all when thinking about San Fran?

Well, a couple of things. Look, we did sell an asset in San Francisco earlier in the year. I wouldn't say that we've given up on San Francisco. I still think that a lot of wood has made San Francisco a really wonderful market for the last decade in the hotel business. A lot of that is still there. It's still a very supply-constrained market generally. It's a hard market to build in when Moscone's up and running and performing very, very well. There's a great mix of leisure, BT and group demand in that market. Again, I think we acknowledge the fact that it's a longer road to recovery than most markets. I think it will continue to trail. But I will say the comparisons to last year get easier and easier. So I think from a growth rate perspective, you'll see some better growth rates year-over-year next year in San Francisco, but probably still lagging pre-pandemic levels more than other markets.

And one moment for our next question. And our next question comes from the line of Michael Bellisario with R. W. Baird.

Jon, just one more on the glamping sort of big picture. Just help us understand how you think about barriers to entry, obviously, lower check size, less dense locations and attractive unit economics, which presumably is one of the reasons why you're doing it, those unit economics, I would think, over time, could possibly narrow with more entrants coming into the space. But just maybe help us understand how you think about barriers to entry and sort of residual value of this type of real estate.

Well, thanks, Mike, and I appreciate the question. Look, first and foremost, I think I'd say that we'd acknowledge this as the most significant risk around the segment and probably the one that we spent the most time, as we evaluated this opportunity getting comfortable with. Look, we believe that the kind of relative lack of barriers to entry are mitigated by a number of things. I think like any hotel investment, we prioritize the best locations in any market we enter. And if you look at where we are at Fredericksburg, which is a market that has a real dearth of quality hotel supply in it, this will undoubtedly be the best location in the market, given its proximity to town and the wineries that are in the region. We believe it's important to have the right product and the right partner. And again, I would acknowledge that anybody can put up a tent anywhere. The products that we're investing in is it's highly unique. It's highly differentiated. It's not easy to replicate. We're partnering with a really strong group that we have spent an enormous amount of time and capital, building an institutional quality technology-driven operating platform that, again, is very, very difficult to replicate. So I do think you'll see supply growth in this segment, to your point, Mike, the unit level economics here are incredibly compelling. But I also think it's important that we acknowledge kind of the point that we're starting at. The going-in yields in this business are in the mid-teens. That's probably double or more what you're seeing in traditional hotel space today. The demand and the revenue CAGRs for the business are projected also to be in the mid-teens for the segment for the next years, again, significantly above, I think, what you'd see in the traditional hotel space. So again, I do expect to see more supply in the business because the unit level economics are so compelling, and I think the growth profile of the business is so compelling. But we are starting for a place that's far, far ahead from a return perspective than we would a more traditional hotel investment.

And then just one more on the same topic. Can you maybe provide some background on your partner kind of what other expertise do they have in this segment aside from this one property? And who are partners and their investors?

This is their first glamping operation. It's been opened for roughly a year. It's obviously performed incredibly well. They have a fairly broad diverse mix of other short-term rental properties, I'd say, generically kind of all across North America. They're -- I think they're a high-quality institutional group. Again, we've looked -- we spent a considerable amount of time trying to find the right partner here. And we do think this is very much the right partner. Again, part of the barriers to entry to replicate what I think that we're going to produce is a lot of the time and the effort and the money they spent building out this technology platform, which is very, very difficult to replicate.

Just to echo on top of kind of what Jon said. If you look -- we've provided a little more information in the investor presentation, but they kind of have an additional 300 locations of [stays] in the alternative accommodation space across 29 different markets. Some of those are urban, some of those are destination markets. And the group there that it's comprised of 3 individuals. They've had a history of partnering with a lot of more sophisticated institutions in the past, MGM, Google Ventures, Greylock and a variety of others. So we view these guys as smart guys. They've been around since 2016 operating in the alternative accommodation space. They navigated the pandemic extraordinarily well, and it was something that we'd spent a lot of due diligence on before we kind of entered into a partnership with them.

And then just one more for me, switching gears. Can you give us some context on the non-same store hotels, particularly Newcrest. It looks like EBITDA sequentially was down like maybe $4 million. How much of that is seasonality? How much of that is maybe ramp up short of your expectations? Any context color there would be helpful?

It's predominantly seasonality, Mike. I mean I think July and August were really hot months in the Sunbelt in Texas in particular. And I think when you look at markets like Dallas and New Orleans, we definitely felt the effect of a hotter summer there. Most of that was expected and underwritten. We've seen, again, a fairly dramatic increase in the performance of the NCI portfolio generally in those markets specifically as we've got post-Labor Day. September and October in Dallas both exceeded 2019 levels for the market. There was some margin pressure in the quarter in that portfolio. A lot of it was driven by the margin pressure that we saw generally across the portfolio in July and August, the higher wages, a more challenging labor market and then just higher utility costs. Again, a lot of that flipped in September when we actually had margin expansion, both in this portfolio and in the broader portfolio over 2019. And I think the operating trends that we've seen in October and what our expectations for the balance of the year, we remain very positive on the portfolio. I think that -- again, if you go back and we've talked about NCI a lot, a lot of the heavy lifting we've done on that portfolio has been done over the last 8.5 months, 9 months. I think we're just now starting to see the benefits of that. And you can see that in terms of how we're gaining market share in some of these markets. And so we're every bit as enthusiastic, if not more, about the opportunity set within the NCI portfolio. Today as the day we closed the deal.

And one moment for our next question. Our next question comes from the line of Austin Wurschmidt with KeyBanc.

So I know the glamping deals are relatively small, but just curious if traditional financing options are available within this segment, both on the construction side, as well as more permanent financing as well as how much leverage do you think you can obtain on a deal on the permanent side?

I would say we're acquiring this first deal unencumbered, and that's obviously based on the size of it. As Jon alluded to, we have been evaluating a pipeline of opportunities that we think are actionable in the near term. And I would say that given the debt capital market backdrop, obviously, financing continues to be challenging for a lot of hotel products that we've actually had some success in talking to regional lenders, who are very familiar with the areas that we're looking -- that are -- where we're looking at potentially developing additional sites. And I would say the quotes that we've gotten in those types of markets have been up to 70% of construction cost, usually for a period of up to five years. So there's an IO period and then a mini term portion to it. But I wouldn't say that the construction financing conversations that we've had around the glamping type of products have been materially any different than you would get with construction conversations around traditional types of lodging product in terms of our experience over the past three months, four months, five months.

And then, also just curious, any parameters that you guys are putting around this segment in terms of either total dollars amount or percent of gross investments until you kind of get a little bit of a better handle on what this looks like through a cycle.

Well, we've obviously started pretty small. And I think we're highly aware that we own 102 hotels worth of about plus or minus $3 billion and one glamping site worth $5 million. And so we'll obviously manage accordingly. We have tried to structure this for some level of growth. And again, I think you can extrapolate over a multi-year, four year, five year period that if we did 10 projects, and they're all $25 million, plus or minus, you can get a couple of hundred million dollars invested in this business. I think what we really prioritize when we structured the joint venture was preserving optionality. And we kind of have 2 points of optionality in every deal that we do. We're going to have the point of optionality when we do the mezz loan and the point of optionality when we exercise the mezz loan. And again, I think our track record using that program, one is it creates this wonderful in-place yield; and two, it does create very valuable optionality. And I think the deal we did in Miami is a terrific example of how valuable that optionality can be because we stepped into a position that's again, probably $25 million to $30 million in the money day one. So I do think we -- again, we started small. The nice thing about this product type is that it ramps up incredibly quickly. I think you're seeing year one yields that maybe aren't fully stabilized, but you're seeing mid-teen year one NOI yield. So the ramp-up to these projects is very, very quick. So again, we have started small, but we've tried to build this pipeline, so we can grow it. And so we, again, preserve optionality on that path to growth.

And then maybe given sort of your comments on this being a labor lite model within your traditional hotel portfolio, what's sort of the latest on labor, what the increases look like going forward and how that sort of plays out relative to demand trends?

Look, I think the labor market is still challenging. I think the way I'd characterize it is it's not getting worse. I would -- and I think that most of the significant wage increase pressures that we've seen are mostly past us. But look, we still have -- it's still hard to hire people. There's still 2x the number of open positions, open positions in this industry that there are available workers. Turnover is still high. The reliance on contract labor is higher than we'd like it to be. It is still very much the number one operating challenge that we have. If you look at our labor expenses relative to what they were in '19, we're actually up about -- we were actually up about 1% in the third quarter versus 2019 levels, and that's with lower occupancy and obviously different cleaning standards. And again, that's driven by a lot of things that I just mentioned, wage growth, contract labor and turnover. So I do think we've done a pretty good job managing through a difficult operating labor market. Our margins for the quarter were down 120 basis points from the third quarter of '19 on a comparable -- other than our same store portfolio. It was about 170 basis points in the comparable portfolio. That's actually a slight improvement from where we were in the second quarter despite the fact that RevPAR was 6% lower. Our EBITDA recapture went up a couple of hundred basis points quarter-over-quarter. So it is still challenging. Again, I think the team is doing a very, very good job managing through that and continues to -- we continue to make progress on the EBITDA recapture.

And then just last follow-up to that, Jon. On the agency labor side, I'm curious what percentage of labor costs that represents today versus maybe what it was in '19 or a more normalized period?

Yes. I mean it's a fairly small part -- piece of our labor. It's about -- it's less than 10% of our overall labor dollars. But on a nominal basis, it's probably double what it was pre-pandemic.

Thank you. And I'm showing no further questions at this time. And I'd like to turn the conference back over to President and CEO, Jonathan for closing remarks.

Well, thank you all for joining today. It's an exciting time here at Summit. We look forward to seeing many of you [Technical Difficulty]. Thank you.

This does conclude today's program. You may now disconnect. Everyone have a great day.