Starwood Property Trust, Inc. Q4 FY2024 Earnings Call

Greetings, and welcome to Starwood Property Trust, Inc.'s fourth quarter 2024 earnings conference call. At this time, all participants are in a listen-only mode. A question and answer session will follow the formal presentation. As a reminder, this conference is being recorded. At this time, I'd like to hand the conference call over to Zachary Tanenbaum, Head of Investor Relations. Zachary, you may begin.

Thank you, Operator. Morning, and welcome to Starwood Property Trust, Inc.'s earnings call. This morning we filed our 10-Ks and issued a press release with a presentation of our results, which are both available on our website and have been filed with the SEC. Before the call begins, I would like to remind everyone that certain statements made in the course of this call are forward-looking statements, which do not guarantee future events or performance. Please refer to our 10-Ks and press release for cautionary factors related to these statements. Additionally, certain non-GAAP financial measures will be discussed on this call. For a reconciliation of these non-GAAP financial measures to the most comparable measures prepared in accordance with GAAP, please refer to our press release filed this morning. Joining me on the call today are Barry Sternlicht, the company's Chairman and Chief Executive Officer, Jeffrey DiModica, the company's President, and Rina Paniry, the company's Chief Financial Officer. With that, I'm now going to turn the call over to Rina.

Thank you, Zachary, and good morning, everyone. Today, we reported distributable earnings, or DE, of $167 million or $0.48 per share for the quarter, and $675 million or $2.02 per share for the year. Across businesses, we committed $1.6 billion towards new investments this quarter and $5.1 billion for the full year, with 67% of our annual investing in businesses other than commercial lending. As a testament to our diverse business model, commercial real estate lending now comprises just 54% of our asset base. I will begin my segment discussion with commercial and residential lending, which contributed DE of $193 million to the quarter or $0.55 per share. In commercial lending, we originated $477 million of loans, which brings our full-year originations to $1.7 billion. Repayments totaled $1 billion in the quarter and $3.6 billion in the year. Our $13.7 billion loan portfolio ended the year with a weighted average risk rating of 3.0, consistent with the prior quarter. In the quarter, we foreclosed on three previously five-rated loans totaling $190 million, all of which were multifamily, two in Texas and one in Phoenix. We obtained third-party appraisals for all three assets, with one indicating a value below our basis. We took a $15 million specific CECL reserve against this $46 million loan prior to foreclosure. In total, the three assets had $61 million of lost sponsor equity. As we work to resolve our existing REO assets, we sold our previously mentioned Portland multifamily asset on our DE basis of $61 million. In addition, subsequent to quarter-end, we received $39 million in repayment of a nonaccrual loan secured by a hospital asset in California that was destroyed in the 2020 wildfires. The insurance proceeds were $1 million in excess of our DE basis. And finally, we are under contract to sell an REO multifamily asset in Texas at our DE basis. Our CECL reserve increased by $36 million in the quarter to a balance of $482 million. Together with our previously taken REO impairments of $198 million, these reserves represent 4.6% of our lending and REO portfolios and translate to $2.02 per share of book value, which is already reflected in today's undepreciated book value of $19.94. Next, I will turn to residential lending. Our on-balance sheet loan portfolio ended the year at $2.4 billion. The loans in this portfolio continue to repay at par with $56 million of repayments in the quarter, and $256 million for the year. Our retained RMBS portfolio ended the quarter at $421 million with a slight decrease from last quarter driven by repayments, and offset by a positive mark to market. In our property segment, we recognized $14 million of DE or $0.04 per share in the quarter driven by our Florida affordable multifamily portfolio. For GAAP purposes, we recorded an unrealized fair value increase related to this portfolio of $60 million in the quarter net of non-controlling interest. The value was determined by an independent appraisal which we are required to obtain annually. NOI for this portfolio increased 9% this year. We expect rents to increase again in 2025 once HUD releases its maximum rent levels in a couple of months. As a reminder, there was a 3.8% holdback from last year that we expect to implement in 2025. Turning to investing and servicing, which we often call Reese, this segment contributed DE of $49 million or $0.14 per share to the quarter. Our conduit, Starwood Mortgage Capital, was the largest nonbank CMBS loan contributor in 2024. During the quarter, we completed five securitizations totaling $595 million at profit margins that were at or above historic levels. This brings our year-to-date total to 17 securitizations for approximately $1.6 billion, the highest level since 2016. And our special servicer, our named servicing portfolio, regained position as the largest named servicer in the U.S., ending the year at $110 billion, the highest level in a decade. This was driven by $5 billion of new servicing assignments in the quarter, and $24 billion during the year. Our active servicing portfolio ended the year at $9.2 billion with $1.5 billion of new transfers, nearly 60% of which were office. In our CMBS portfolio, we purchased a $49 million B-piece. And in this segment's property portfolio, we acquired $14 million of property investment. Concluding my business segment discussion is our infrastructure lending segment, which contributed DE of $22 million or $0.06 per share to the quarter. Our strong investing pace continued this quarter with $532 million of new loan commitments bringing our total for the year to $1.4 billion, its highest annual level to date, with this performing loan book ending the year at $2.6 billion. And finally, this morning, I will address our liquidity and capitalization. During the quarter, we executed $2.3 billion in debt transactions, which Jeff will talk more about. We repaid our $400 million December unsecured at maturity, and early repaid half, or $250 million, of our March 2025 high yield maturity. After repaying the remaining $250 million next month, we will have no other 2025 maturities, and our next corporate debt maturity is not until July 2026. Our liquidity position remains strong at $1.8 billion. This does not include liquidity that could be generated through sales of assets in our property segment, direct leveraging of our $4.9 billion of unencumbered assets, or issuing high yield or term loan B backed by these unencumbered assets. We continue to have significant credit capacity in our business lines, with $10.5 billion of availability under our existing financing lines, and unencumbered assets of $4.9 billion. Our leverage continues to remain low with an adjusted debt to undepreciated equity ratio of just 2.1 times, its lowest level in over four years. With that, I will turn the call over to Jeffrey.

Thanks, Rina. We ended 2024 with a flurry of capital markets transactions where we extended the average term on our corporate debt from 2.2 to 3.5 years, repriced upside and extended $1.4 billion in term loans, extended and upsized our corporate revolver, and issued high yield unsecured debt, all at the tightest floating rate spreads in our company's history. We raised almost $800 million in incremental proceeds in this process, leaving us with significant investable firepower as we enter 2025. We've seen significant spread compression across our investment cylinders in CRE, RMBS, fifth, residential loan and financing spreads, and in CRE cap rates, where spread tightening has offset most of the rate rise since the election. Liquidity has returned to all of these markets for both new transactions and the refinancing of the record loan originations volume from late 2020 through early 2021, giving us as strong a loan pipeline as we have seen in three years. Banks continue to earn significantly higher ROEs lending to nonbank lenders like us than making their own direct whole loans to borrowers. They've also reduced the lending spreads to us in line with the spread compression I just spoke about, allowing us to compete at tighter spreads across our platform on higher quality, low-leverage loans at today's lower basis, which we believe will create outsized opportunities for us in 2025. We've already closed $1.5 billion of loans in the first quarter and our business plan for 2025 is to write the most loans we have in any year since our inception other than 2021, which was a record year for lenders across the board. Our unique diversified business model has a cyclical debt to equity ratio of 2.1 times today, leaving us significant room to invest our near-record cash levels while still maintaining our low leverage business model. Investing an incremental billion dollars of equity will offset the drag created by our REO and non-accrual balances today, and given we can borrow today at record low spreads, allowing us to more than absorb tighter lending spreads. Our approximately 350 employees at Starwood Property Trust, Inc., in addition to the employees of our manager, Starwood Capital Group, are working toward this goal of significantly increasing our investing pace across the platform. Although we have a lot of work to do, we believe we will be able to start exiting legacy non-accrual and REO assets at a faster pace. We presented our board with our 2025 plan this quarter and in that, we have a plan to reduce this portfolio, which has caused significant drag on earnings, by half in 2025 then by half again in 2026 as we look to exit our difficult legacy positions by 2027. While we have earned our dividend in these difficult years where CRE was the hardest hit by the Fed's unprecedented interest rate increases, freeing up this trapped equity while simultaneously taking advantage of tighter spreads to grow our investment portfolio will allow us to increase earnings in the coming years while holding on to the vast majority of our over $1.5 billion in unique harvestable gains in our own real estate portfolio. We had $3.6 billion in repayments in CRE, and with near-record liquidity and access to capital, we have approached resolutions differently than most of our peers, focusing on the highest NPV to shareholders after factoring in the workout timetable and cost of capital of exiting loans today versus improving performance and waiting for a better exit window. We work with our manager, Starwood Capital, with over $110 billion under management across nearly all CRE asset classes, to find the most accretive business plan for difficult assets. If you'll recall, back in 2021 and 2022, we made $93 million for shareholders after foreclosing, holding, and repositioning our first defaulted loans, two former Winn Dixie industrial assets. This quarter, we sold our Portland REO Multifamily at our basis. Looking to Q1 of 2025, we will sell an REO multifamily in Texas at our basis. And we already were repaid a hotel in Napa that burned down at $1 million above our basis. All of these were as we expected and signaled. We are beginning interior demolition on a $115 million office building we've foreclosed on in D.C. and are converting it into a beautiful multifamily, which you will see on the cover of our supplemental. Rather than sell this asset in a depressed D.C. office market, our underwriting has us returning a gain to shareholders upon completion. We have resolved or modified 25 assets totaling $2.8 billion to date, with 12 modifications and 12 assets we have taken into REO to reposition or sell as we did in the previous example. With markets repairing, we expect the pace of resolution to pick up going forward. Rina mentioned our significant liquidity, and I will add that we are through the vast majority of our projected deleveraging. Our four- and five-rated loans, which were optimally levered with $794 million of repo and $901 million as CLO debt, today have only $144 million and $620 million of debt, respectively, down 81% and 31% on the same asset base. Having paid down 81% of our repo borrowings already on these fours and fives, we are left with only $144 million of repo debt subject to any potential margin calls remaining on these assets. As a result, we have significantly more clarity into our future liquidity than we have had at any point in this higher rate cycle, which gives us comfort that now is the time to go fully on offense with our incremental liquidity and access to significantly more liquidity. As Rina just mentioned, we will also use tethered spread to continue to increase our unsecured corporate debt as a percentage of our overall debt and we'll use proceeds to continue to pay down secured asset-level debt, which we believe will put us in discussions with our rating agencies to upgrade our corporate debt ratings, thereby further reducing our cost of capital and making our planned growth even more accretive to shareholders and keeping us on our stated path of getting to investment grade. In commercial real estate lending, although we've foreclosed on three multifamily loans, our four- and five-rated loans were down this quarter. We have said in the past that late cycle multifamily loans will be the hardest hit by stubbornly high forward rates as undercapitalized borrowers will struggle to buy new caps and extend loans. That means more work for our asset managers; our manager is one of the largest owners of multifamily in the country with 106,000 units and we expect, as we have done in the past, that we will quickly improve performance and be able to exit these assets at or near our basis, as we have done multiple times recently, or choose to hold them as accretive long-term investments. As I mentioned, financing costs continue to improve and our pipeline is as big as it's been in years. We plan to add to our personnel again this year to take advantage of what we see as one of the best new lending environments we've seen. In infrastructure lending, CLO and borrowing spreads continue to move down, offsetting tighter lending spreads, and after making $1.4 billion in accretive investments in 2024, we expect to grow at a much faster pace in 2025, given tailwinds in the energy sector and the much-discussed need for incremental power in our country. We're off to a strong start, with $229 million closed and $763 million in the process of closing, all a blended ROE in excess of what we can earn in our core CRE lending business. We hope and expect this business will continue to grow and become a bigger part of our asset base going forward. In our real estate investing and servicing division, as Rina mentioned, we are now named special servicer on $110 billion of CMBS loans, our most since 2015, the most in the growing CMBS market. In a slow CRE lending year, Reese was a strong contributor to earnings in both special servicing and our CMBS conduit SMC, highlighting once again the benefit of our multi-cylinder platform and the positive carry credit hedge embedded in our business as our servicer will again make more money as the workouts of loans made in a lower interest rate environment continue to come through in the coming years. With that, I'll turn the call to Barry. Thank you.

Thanks, Zachary. Thanks, Rina, and thanks, Jeffrey. This is going to be one of the more interesting calls I've had in ten years. I want to talk about the macro and why the windshield has never been murkier. Nobody really knows the effect of tariffs or if they're going to be targeted or broad. But there's one short-term conclusion, which is that tariffs are inflationary. There are only three places price increases can go: they can go to manufacturers, to consumers, or be split between the two. Take steel tariffs, for example: American producers will raise prices something less than a 25% increase on foreign imports. We'll have to wait and see. I think for the last nine months, 99 out of 100 economists would have thought with our deficits the 10-year Treasury was going to four, five, or five and a half percent. And yet we sit this morning with the 10-year at about 4.28% in the context of a huge rally, which shows the inherent weakness in the U.S. economy. What you're seeing is that ten percent of the population is spending half the money, and the bottom half is not participating because technology, including AI, has led to concentrated benefits. AI has driven massive capital commitments from a handful of companies that have leveraged significant sums. It is similar to an infrastructure bill in scale but being spent much faster. That's the exceptionalism of the U.S. economy. Looking at education and the school system, we have a distorted economy and consumer sentiment is falling because of political and policy uncertainty coming out of the White House. We'll see if it's tactical or not; every day there are new developments and we have to readjust our views. What this means for real estate is interesting: we're in a good place. Construction has come down — you've heard it from peers: multifamily starts down 60-70%, industrial starts down 70%. Buying real estate today with today's interest rates often means buying below replacement cost. That implies rents must rise to justify new construction in many cases, which is bullish for loans we have in place and for existing assets. With tariffs, lumber will be more expensive, steel prices rise, and construction costs go up, which leads to future inflation. If multifamily supply is very constrained, rents will rise. For example, in Denver completions went from 17,000 in a quarter to 3,000 and rents moved to double-digit growth. Those factors feed into CPI and may show up in late 2026 or 2027. So we're caught: we want low rates to refinance our debt — great for the real estate complex, LTVs, and cap rates — and we also don't mind higher rates because we can lend at higher spreads and make more on the new book. It's a tug of war. We don't mind higher rates, but one reason we're busy is that rates are stable and markets expect them to remain so. The market's view can change quickly, of course. Nobody really knows today; it's been soft and two weeks ago it was a runaway freight train, so markets and companies are confused and this complicates capital spending decisions for average companies. While we applaud the concept of rebuilding manufacturing in the U.S., it's hard to instantly produce manufacturing jobs if the workforce isn't available. Add that to potential large deportation measures recently discussed in Congress, which could put pressure on wages, and many of those people work in construction. That will increase construction costs and contribute to a lack of rebound in construction. Those are the jobs like Uber drivers, construction, landscaping, agriculture — the backbone of many local economies. The markets are clearly worried, but right now the 10-year doesn't indicate an imminent surge. Trump's policies by consensus would increase the deficit, especially if the Senate blocks spending cuts the House has proposed. Now, I want to talk about us because I think the company is in fantastic shape — probably the best it's been in years, even with our non-accrual loans. Look at the balance sheet: 2.1 turns of leverage, down from nearly three. We can easily borrow money, as Jeff said, and increase our leverage and earnings power. We're going to be aggressive in growing our lending book. A few details we didn't highlight: our construction exposure was 24% of our loan book at one point and is now down to 3%. Forward fundings under those loans were 34% of our lending book then; today it's 8%. We don't have meaningful forward-funding risk. We've delevered repos significantly and future funding required for any margin calls would be de minimis in the context of the firm. The company is on a rock-solid foundation with $1.8 billion of liquidity and maybe better than we've ever been. I'm excited to see our businesses like the conduit do 17 securitizations this year — that's more than one a month — which shows we're a manufacturing facility for paper: take it in, package it, and sell it. LNR is now again the largest servicer in the nation, which should be good for future earnings from that subsidiary. Those businesses are unique to us in the mortgage area and have enabled us to be the only mortgage REIT not to cut the dividend over the last cycle. When we foreclose on multifamily, I tend to like owning those assets because we land at about 65% of cost typically, and by definition we're below replacement cost. Multifamily markets will stabilize as new supply is absorbed. It won't happen this year; there's still too much supply coming this year, but it should fall off a cliff in 2026, and buyers are positioning themselves. Multifamily cap rates have tightened by about 75 basis points already and may tighten further as future rent growth becomes more apparent. Because LNR is so large, we have a front-row seat to restructurings and a proprietary deal pipeline. That allows us to work with borrowers and offer solutions, whether preferreds, seniors, or equity. I think we'll pick up our residential lending business again, a vertical that's been largely closed for some time; we'll look at home price appreciation and decide how much capital to deploy. Our energy business (SIF) is something I really want to grow dramatically; it's been a gift that keeps on giving. Loans are being paid off quickly, and spreads have compressed, but overall borrowing costs are roughly where they were. The CMBS market is wide open — we had one of the biggest issuance days in history two weeks ago — and you can refinance pretty much anything in CMBS, including office. That's good for us because we have some exposure to office. We're doing our first massive data center loan; we expect to do two more. These are more infrastructure-like loans: generally to the best credits in the world with long leases and attractive debt yields. We expect that business to grow. As we add more diversity to the portfolio and have made it through the five-hundred-basis-point rate hurricane the Fed threw at us, convincing the rating agencies to give us investment-grade would be meaningful. We're a couple of notches away but have the highest credit among our peers in the REIT world, and getting upgraded would reduce cost of capital and help growth. One business we've put on pause but are likely to come back to is buying equity real estate. Our affordable housing portfolio is a prime example: nearly a $2 billion gain with no net equity invested today — it's a gift that keeps on giving with strong trailing rent growth and forward rent forecasts. We can do three to four billion dollars of investments without issuing equity, simply by increasing leverage to more normal levels. We're underlevered today relative to our historical norms, but we don't need to borrow at the moment. If the pipeline gets really big, we'll lever up to more normal levels and still remain conservatively capitalized. The team is intact, working hard, and we're ready to go on offense. We're already losing deals because it's becoming competitive again, even for office construction loans. There are many private credit players entering real estate; because of our scale, we get big phone calls for large mezzanine pieces where few counterparties can participate. We're pleased with our position. As we resolve non-accrual loans, that capital comes back and can be redeployed. The goal is to be thought of as a finance company in REIT form — a company with multiple business lines paying out earnings efficiently — and we'll continue to add business lines. We've looked at buying some distressed lenders and other businesses to grow other lines; those opportunities are being evaluated. With that, thank you and I'll pass the call back to the operator.

Thank you. At this time, we'll be conducting a question and answer session. Star one on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star two if you like to remove your question from the queue. One moment please while we poll for questions. Our first question comes from Stephen Laws with Raymond James. Please proceed with your question.

Hi. Good morning. I see a few topics I want to hit on with WoodStar. First on the expense side, cost to run operations was a little higher in Q4. Looks like that's been the case the last couple of years. Is that seasonal, and how do we think about that moving forward? And then on the interest expense, what's the remaining term on that debt? And how do you think about what that's going to cost when you look to refinance that? And then, as I think about the fair value mark, typically we've seen that take place in the second quarter around those annual rent increases. How do I get my hands around what drove that valuation gain this quarter and how should I think about forecasting that as we move forward?

Yeah, thanks, Steven. Appreciate it. We have two and a half years remaining on that debt, and we will be opportunistic — we'll go early if the market gives us an opportunity as we always have. Two and a half years is plenty of runway. Expenses were really up due to hurricane-related maintenance that needed to be done, and that is not run rate, so we expect expense to come back to run rate. As Barry said, with eight percent rent growth and probably another eight percent coming next year with moderating expenses, we feel pretty good. On the fair value mark, it can be difficult to look at the quarter where the 10-year moved up in the fourth quarter; I'll note it's down 27 basis points since year-end. In the first three quarters we use desktop underwriting, and in the fourth quarter every year we get an appraisal. The appraisal uses a discounted cash flow method. It backs into an asset-level cap that translates to roughly a 4.68% cap in our portfolio. I have a list of the last twenty trades in this sector in Florida in our markets, and the last twenty of them for significant size have a 4.6% blended cap. Cap rates have come down and I would expect they will continue to come down, certainly as we look at our desktop mark for next quarter. The appraisal looked at rent growth for this year and we effectively get half of that because it's for six months; a 3.8% holdback that's been recognized is worth about 19 basis points in cap, and on a six-month basis it's equivalent to about ten basis points. So the 4.68% asset-level cap used in the appraisal is almost a 4.78% equivalent. We're about 18 basis points above the average of the last 18 months, which should have come down, so while the appraisal was tighter than our desktop, we feel comfortable that it is within the market range today at worst.

Our next question comes from Rick Shane with JPMorgan. Please proceed with your question.

Hey, guys. Thanks for taking my question this morning. One of the anomalies in the market is that you guys are trading at a significant premium to virtually all of your peers. That creates an interesting arbitrage in terms of acquisition. When we normally raise this question, company responses are, 'Why would we want to buy anybody else's problems?' But given your experience in special servicing and thoughtful analysis of loans, does it make sense to scale the business at this point by inorganic opportunities as well as organic opportunities?

We're agnostic. Any business plan that needs a return of capital and is accretive, we'll look at. We don't really have peers in the same way. We pay a sizable dividend and given the diversity of our business model and its historic ability to cover that dividend over many years, we should trade at a lower dividend yield than where we are trading today. We have businesses that generate fee-based revenue like our manager and LNR, and we should trade at a premium to peers. We are mispriced in my view. We've traded around 1.2x historically. If we were back at 1.2x book, the market capitalization would be several dollars higher. We have businesses that will perform on an ROE basis and the ability to grow book value internally. Our WoodStar portfolio — our affordable housing portfolio — will keep appreciating in value for the next handful of years at a minimum. I think we're completely mispriced and there's an argument that we should be trading at a premium, given our diversification and earnings power, but that determination is for the market. We're going to grow the business significantly this year and try to prove that out.

We'd love to acquire other companies at a discount, but sellers don't often want to transact that way. We'll grow as fast as we can organically. Ten percent of our assets are in U.S. office and our peers often have significantly higher exposure to office, which helps explain differences in valuation. In our view we should trade at a premium.

Our next question comes from Jade Rahmani with KBW. Please proceed with your question.

Thank you very much. Life science has been an area of significant challenge due to oversupply and the basis many of these projects are on. Many spec projects are being evaluated for conversion to office because that's the cheaper option, but rents per square foot in office are much lower. Could you discuss the one life science downgrade you experienced and what the outlook is there? Also, on new initiatives, GSE multifamily is a business you've been historically interested in. Is there potentially a move in that direction without making a huge bet — perhaps a joint venture — that you might be interested in?

We never really leaned into life science. We did one loan under $100 million in Boston in a great Seaport location. We like our basis relative to today's lower rates; if we can sign a lease near $90 per foot, which we believe the market is around $92, we would be out of that loan, but we need to find that lease. Life science is experiencing a difficult run due to supply. To some degree, AI could reduce labor or experiment iterations in life sciences, which may reduce demand for lab space over time. We have the least exposure to life science of our peers. The basis is high and conversion back to office is difficult; if converted the rents likely won't return equity and might not return a full loan balance. So it's a difficult sector. On GSE multifamily and DUS lending: we'd love to own one. We have tried several times to buy one but licenses and broker relationships are complicated and expensive, often requiring multiyear guarantees and distribution networks. If there's privatization, the landscape could change. JVs are an option but they often come with underwriting we don't like. We've looked at them and in many cases we find other places to deploy capital that are more compelling today: energy infrastructure, growing our CRE lending, and restarting property investments. We'll continue to look at DUS opportunities and JVs, but only on our credit terms, not someone else's underwriting box.

Our next question comes from Doug Harter with UBS. Please proceed with your question.

Thanks. You talked about expanding owned property investments and getting back into buying properties. Do you think that would come with selling down some of the existing assets, or would that just be deploying some of the excess liquidity you have today?

We would use excess liquidity to buy properties; we would not need to sell existing assets to fund purchases. We have about a billion of cash and additional availability, so for the foreseeable future we can deploy without selling. If great opportunities arise and people approach us, we might sell some assets opportunistically, but we don't need to. We've made strong gains in REO historically, and we also look at possibilities like triple net or other property investments where credit is high. We could make those work accretively, though the market tightness on high-credit assets makes pricing challenging. Our capital team is looking at opportunities and will continue to show ideas. That's a natural place to add exposure.

Our next question comes from Donald Fandetti with Wells Fargo. Please proceed with your question.

Hi. Can you talk about the Washington D.C. office and multifamily market? More people are coming back to work, but potentially fewer people live in the core. How are you thinking about that area, given your assets there?

We have one building in Virginia and two in D.C., including 1201 which is a conversion target. It's hard to predict the outcome for D.C. office given current political and federal employment dynamics. Return-to-office trends are positive for retail and building-level amenities, but the churn in federal staffing could affect leasing, particularly GSA leases. Our largest D.C. loan is 84% leased and our other D.C. loan had another re-up recently, so those are not terrible. Our Virginia asset is smaller and around 64% leased, which is more challenging. We have limited margin-call-eligible repo on our lending book and have paid down most of those exposures, so a deterioration wouldn't be a liquidity problem but could extend exit timelines. We need to see more clarity on government leasing and return-to-work patterns to better gauge the market recovery.

We've reached the end of the question and answer session. I'd now like to turn the call back over to Barry Sternlicht for closing comments.

Thanks everyone for joining us today, and good luck in this fascinating environment we're in, and we wish you well. We'll see you next quarter.

Thanks.