TRTX - NYSE

Good morning, and welcome to the TPG RE Finance Trust Second Quarter 2023 Earnings Conference Call. Please note this event is being recorded. I would now like to turn the conference over to Deborah Ginsberg, General Counsel, Vice President and Secretary. Please go ahead.

Good morning, and welcome to TPG Real Estate Finance conference call for the second quarter 2023. I'm joined today by Doug Bouquard, Chief Executive Officer; and Bob Foley, Chief Financial Officer. Doug and Bob will share some comments about the quarter, and then we'll open up the call for questions. Yesterday evening, we filed our Form 10-Q and issued a press release and earnings supplemental with the presentation of our operating results, all of which are available on our website in the Investor Relations section. I'd like to remind everyone that today's call may include forward-looking statements, which are uncertain and outside of the company's control. Actual results may differ materially. For a discussion of some of the risks that could affect results, please see the Risk Factors section of our 10-Q and 10-K. We do not undertake any duty to update these statements, and we will also refer to certain non-GAAP measures on this call. And for reconciliations, you should refer in the press release and our 10-Q. With that, it's my pleasure to turn the call over to Doug Bouquard, Chief Executive Officer.

Thank you, Deborah. Good morning, and thank you for joining our call. Over the past quarter, the market has begun to reflect a greater livelihood of a soft landing as the Federal Reserve has been successful in dampening inflation without curing a recession. Over the past 16 months, front-end interest rates have increased from nearly 0 to their highest level seen in 22 years. Despite the Fed's restrictive policy in place, we should acknowledge a few noteworthy takeaways. First, the labor market remains incredibly strong; second, the residential housing market has remained resilient and positive national price appreciation despite mortgage rates hovering around 7%; third, credit spreads in the corporate credit market have tightened dramatically and have almost fully retraced to their pre-Fed hike levels; and lastly, the U.S. public equity market has rallied nearly 18% year-to-date at the S&P 500 quickly approaches an all-time high. Despite the broader positive shift that in risk sentiment in many asset classes, commercial real estate continues to underperform. This underperformance is driven by the pressure on values across the commercial real estate landscape, reduce liquidity for debt and equity and secular challenges facing the office market. Ultimately, there are a variety of micro and macroeconomic forces that drive commercial real estate price. But in addition to the headwinds I just mentioned, continued uncertainty related to both spot and forward interest rates continue to create volatility on valuations and the hesitation on investment activity across the sector. Our view remains that pressure within the commercial real estate space may persist for an extended period. And therefore, we continue to maintain our strategic posture of, first, cautiously deploying capital; second, maintaining a strong liquidity position; and third, proactively risk managing our investment portfolio in a disciplined and tactical manner while focusing on long-term maximization of shareholder value. Over the past quarter, our decline in net income to shareholders was driven primarily by a $56 million net increase in our CECL reserve, which reflects our view of sustained value declines within the office market. Despite the net increase in our CECL expense, we continue to make progress reducing exposure to credit challenged assets. In May, we sold a $71 million Brooklyn office loan. And post quarter end in July, we sold a $129 million office and retail loan located in [indiscernible]. For each of these loans, we determined the optimal resolution path was via loan sale and executed accordingly. It's also worth noting that TRTX provided no seller financing as part of these transactions. And on a combined basis, the recovery amounts exceeded the carrying value, at which we held them on our balance sheet. For broader context, over the past 15 months, we have reduced our cumulative office exposure by approximately $1 billion. From a balance sheet perspective, we fund no new investment this quarter. However, we did see -- we did receive $279 million of repayments during the second quarter, primarily from multifamily refinancing activity. While the multifamily sector has experienced cap rate widening over the past year, both debt and equity capital remain available, albeit at a lower entry point relative to the peak of the cycle in 2021. From a liquidity perspective, we continue to maintain a defensive posture. Total liquidity exceeded $542 million, broken out as $307 million of cash; $206 million of CLO reinvestment capacity; and $28 million of undrawn capacity on our secured credit facilities. Our performance this quarter continues to reflect an asset management strategy that relies on the depth and breadth of TPG's broad real estate platform with approximately $20 billion of AUM across the mix of debt and equity strategies. From an asset resolution perspective, we continue to maximize value for our shareholders regardless of whether we execute a loan modification, sell a loan or for close on asset. We continue to believe the kick-to-can approach for assets facing long-term secular headwinds is not a winning strategy. Furthermore, our deep investment experience across multiple business cycles and diverse -- and our diverse real estate investment platform allows us to evaluate and execute on all potential resolution strategies. Inherent in this approach, the market should expect a certain amount of near-term volatility in our distributable earnings, and this quarter was no different. However, we continue to have high conviction that this is the optimal long-term strategy for our shareholders and will position TRTX for growth as the real estate credit market continues to evolve. With that, I'll turn the call over to Bob to discuss our financial results.

Thanks, Doug. Good morning, everyone. GAAP net loss to common shareholders for the second quarter was $72.7 million compared to GAAP net income to common shareholders of $3.8 million last quarter. The change was due primarily to an increase in CECL expense of $81.3 million, reflecting our assessment of further weakening in liquidity for and valuations of office properties nationally. Net interest margin for our loan portfolio was $26.1 million versus $21.7 million in the prior quarter, an increase of $4.4 million or $0.06 per common share. Distributable earnings declined quarter-over-quarter to a loss of $14 million versus earnings of $13.4 million in the prior quarter. This change was due primarily to realized losses of $33.2 million related to two office loans resolved during the second quarter, one by sales and one by conversion to real estate owned. These realized losses approximated the CECL reserves . Distributable earnings before realized credit losses was $19.1 million or $0.25 per share, a 43% increase from $13.4 million or $0.17 per share in the prior quarter. Book value per share at quarter end was $13.10, a decline of $1.21 compared to the first quarter of this year due to CECL expense in the second quarter of $1.15 per share. Our CECL reserve increased quarter-over-quarter by a net amount of $55.9 million to $278.3 million from $22.4 million as of March 31. Our CECL reserve rate increased to 572 basis points from 420 basis points due almost entirely to further weakening in the office market. At quarter end, our five individually assessed loans represented 63% of our total CECL reserve. Regarding liquidity, we maintained high levels of intermediate -- excuse me, of immediate and near-term liquidity, roughly 10.9% of total assets to support our loan investment and asset resolution strategies. Cash and near-term liquidity at quarter end was $542.9 million, comprised of $307.4 million of cash, $206.7 million of CLO reinvestment cash and $28.7 million of undrawn capacity under our various secured credit agreements. Our third CLO remains open for reinvestment through February of 2024. This term, non-mark-to-market, nonrecourse financing, with a credit spread of 202 basis points, is valuable in supporting new loan investments, optimizing our current financing arrangements and maintaining a high proportion of non-mark-to-market financing. During the quarter, we funded $33 million of deferred funding commitments. Unfunded commitments under existing loans declined by $53.4 million to $300.5 million, which is only 6.2% and of our total loan commitments. Regarding credit, risk ratings remained unchanged at 3.2. Nonaccrual loans at June 30 totaled $555.6 million or 12.2% of aggregate UPB versus $558.9 million or 11.3% of the aggregate UPB for the prior quarter. The de minimis net change quarter-over-quarter masks important asset management activity during the quarter. We reduced nonaccrual loans by $126.3 million through the resolution of two office loans during the quarter, one by sale and one by conversion to REO. Nonaccrual loans increased by $129.2 million related to a 5 rated first mortgage loan secured by an office and retail property in Manhattan that went nonaccrual during the second quarter due to a maturity default. That loan was promptly sold in July. Consequently, nonaccrual loans today totaled $426.4 million. Nonaccrual status is the logical and often necessary consequence of our asset management steps to resolve loans through Patriot Capital and thoughtfully allocated to create shareholder value. We work to resolve nonaccrual loans promptly to maximize recovery and boost shareholder value. We realized losses during the second quarter on two 5-rated defaulted first mortgage loans, both on office properties. $24.1 million or -- excuse me, or 33.8%, other loans $71.3 million UPB on a 444,000 square foot former warehouse building converted to office in Brooklyn. We sold the note after a broad marketing process conducted by a national brokerage firm. We sustained a $9 million or 16.2% of the loan's $55.7 million UPB on a 375,000 square foot office building in downtown Houston. We converted that nonperforming loan to real estate owned after conducting an extensive loan sale process, which led us to conclude, shareholder value will be maximized through ownership skilled property management and leasing using TPG Real Estate's broad resources and experience and eventual sale. We borrowed $31.2 million under a 5-year, 7.67% fixed rate interest-only mortgage loan that is assumable, which we believe may facilitate our eventual sale of the property at the appropriate time. The building is 77% leased and occupied. Based on current operating results and mortgage financing, our investment generates an 8% return on cost and a 10% levered ROE after depreciation and amortization expense. And in late July, we sold a $129.2 million first mortgage loan secured by an office and retail property in Manhattan. That loan sale is disclosed as a subsequent transaction and will be reflected in our results of operations for the third quarter. Regarding CECL, our CECL reserve increased this quarter by $55.9 million net, reflecting a lack of debt capital for office transactions, which has pressured valuations. Our general reserve declined by $56.3 million, largely due to the transfer of 3 loans from the general pool to the specifically identified pool. Our specifically identified reserve increased by $112.2 million due to increased loan loss estimates and the aforementioned transfers. Regarding our capital base, our focus remains on maintaining a sturdy diverse financing base. Our deep relationships in the financing community and the market power of TPG's Firm-wide Capital Markets business enables us to access and maintain long-dated, cost-efficient debt capital to support our existing portfolio and selected loan purchases and originations. During the second quarter, we used $265 million of reinvestment capacity in FL4 to finance an equal amount of existing loans and retire $189.8 million of borrowings under five different secured financing arrangements. Quarterly interest expense savings are approximately $0.05 per share. At quarter end, we had $206.7 million of reinvestment capacity available in FL5 for similar use or to support new loan acquisitions or originations. We also extended for 1 year on existing borrowing arrangements with Morgan Stanley with $500 million of committed capacity and $54 million of borrowings at quarter end. We closed a new 3-year, $200 million credit mark only secured financing arrangement to Bank of America with $35.9 million of borrowings in June 30. And simultaneously, we allowed to expire a $200 million mortgage warehouse with BofA that had been in place since 2018. [indiscernible] test of our financial covenant package, which is uniform across all 12 of our secured borrowing arrangements and reduce the threshold to 1.4x from 1.5x. We saw this change to accommodate higher benchmark interest rates and elevated levels of nonperforming loans as we continue to execute our asset resolution strategy. This quarter, we've already delivered notice to exercise our buy right extension of an existing $500 million credit facility with Goldman Sachs with current borrowings of $324.3 million. At quarter end, 71.7% of our secured financing was nonrecourse, non-mark-to-market versus 74.1% at prior quarter end. These levels are consistent with our long-standing strategy relying primarily on non-mark-to-market, nonrecourse term funding. Our debt-to-equity ratio declined quarter-over-quarter to 2.79:1 from 2.95:1, and we were in compliance with all of our financial covenants at June 30. And with that, we'll open the floor to questions. Operator?

[Operator Instructions]. Our first question comes from Stephen Laws with Raymond James.

Bob, I guess you touched on, I believe, the sale post quarter end of the 5 rated mixed-use loan. How much of an impact will that have in Q3? And how much of that is an impact to GAAP versus what's going to run through distributable earnings?

Sure. Good question. That transaction is part of third quarter results. It will be reported in October, but -- so there will be a loss. It will flow through distributable earnings. The loss we're comfortable is covered by the existing CECL reserve.

Okay. So expect it to be covered, but no material impact to GAAP then?

That's right.

Great. Touching base on FL5, looks like about a little less than $200 million of reinvestment capacity that's, I guess, available through, I think, December is when that window end. Can you talk about how you look at replenishing the collateral there and thoughts about when that might happen? And how much net interest income that might add when it's done?

Sure. This is Doug. I'll answer that. So in terms of our liquidity, we've maintained this CLO reinvestment capacity really continuously over the past year. And really what it allows us to do is have a certain amount of optionality in terms of potential uses of capital. In terms of how we think about deploying it, Bob can kind of walk through in more granular detail exactly on -- from a per share basis. But we expect to be deploying that over the next 3 to 6 months, generally speaking, in advance of that booking window open. So we will updated in term of our process and our prior end in terms of deployment of capital.

And to be specific about the timing, FL5 has a reinvestment period of 2 years, concludes in February of 2024. There's actually effectively a 59-day tail on that. You'll recall with respect to FL4, as reinvestment period closed this year in March that we undertook a similar reinvestment strategy, which I described earlier on the call that concluded in May. The rough math there is if we were to do the same thing with FL5 as we did with FL4 and use that capacity to finance existing loans, which is only one of multiple paths as Doug just described. . And you assume that existing loans are borrowed is around 75% under other forms of financing. If we had $207 million of cash in FL5 that would equate to about $156 million of repo borrowings, which we would repay. That costs a little more than $200 million over term SOFR. So that all works out to be a little bit -- it's basically between $0.04 and $0.05 a quarter of interest savings that would result from FL5 reinvestment.

Our next question comes from Rick Shane with JPMorgan.

I've been bouncing around, so if this was covered in the prepared remarks, I apologize. Can you talk about were there any buyouts from the CLOs during the second quarter related to defaults? And as you look through the third quarter and what's going on in the CLOs anything that you anticipate buying out?

Sure. With respect to the first question, Rick, no, we didn't buy any loans out of any of our CLOs in the second quarter. With respect to what may happen in the future, that's really a function of loan performance, which we attempt to predict but no one can predict with perfect precision. It's also a function of the cushion that's structured into the over collateralization and interest coverage tests each of our CLOs. For those who have studied our CLOs carefully, you'll see that, for example, in FL3, which is the most mature of our three outstanding CLOs, we actually have a very substantial over-collateralization buffer there. So that transaction could actually withstand loan defaults for loans that are in that structure and would remain in that structure without triggering a diversion of cash flow.

Got it. And just to be clear, and I believe I know the answer to this, but I just want to make sure, when you talked about amending the coverage test that was within -- that was across your secured credit agreements, correct?

That's correct. So it has no impact on the CLOs. It's only with respect to our credit facilities and note-on-note of other non-CLO financing arrangements.

Our next question comes from Sarah Barcomb with BTIG.

So we're expecting to see some distributable earnings volatility, as you spoke to on that loan sale in Q3 like we saw in Q2. But can you talk about where we sit now with respect to the watch list? Should we expect to see additional watch list migration and higher specific loss reserves going forward?

Sarah, I appreciate the question. So first, I'll just address our migration in terms of the 5 risk-weighted asset bucket and sort of take those, the specific assets and generally put it in two categories. The first category, as Bob and I previously mentioned, relates to the subsequent event disclosure in connection with the sale of the office and retail loan located. So call that category 1, which has been resolved. And to Bob's comment, we'll be sharing more information on that loan sale at the conclusion of our third quarter. . The second bucket, which leads four 5 risk-rated loans, really we are currently pursuing various resolution paths, including potential loan sales within that bucket. And those assets specifically are all office assets across four different office markets: Philadelphia, New York, Virginia and Orange County. And given that we're actively in the market pursuing various or modules of providing transaction specific information at this time. But I think what I would highlight as it relates to our 5-rated assets and our resolution of credit challenged loans, I think it is worth highlighting that in the three loans that were most recently resolved, including our subsequent event, which we've disclosed in the aggregate across those three resolutions, they're resolved in excess of our carrying value. So I continue to kind of point to that in terms of both the comment on our CECL reserve, but also a comment on our ability to effectively resolve credit channels and assets on our balance sheet.

Okay. So yes, I mean you've been managing some in-place issues on your book, but you also originated a hotel loan after quarter end. So I was hoping you could speak to how you're thinking about maintaining more defensive liquidity versus going out and originating loans at higher coupons? Should we see new originations pick up in the second half of the year? Or how do you think about that?

That's a fantastic question. I think as we continue to make progress through our credit challenged loans, you will see us begin to more actively deploy capital in terms of new originations. I think that, again, as you look at our current bucket of 5 risk-weighted loans as we make progress, we look forward to the ability to resolve and then recycle that capital. That's been a pretty consistent strategy of ours that Bob and I had mentioned in prior quarters, which is our core view relates to the fact that we don't think that a kick-to-can strategy for assets that we see having long-term secular pressures is going to be a winning strategy for our company. So we've been really disciplined around addressing issues, being overly transparent with the market. And then as we address those issues, we'll be very actively engaging in the new origination market. Just for one moment in terms of capital deployment, I think that what we are seeing is actually a very ripe opportunity to be lending. I think despite the fact that broader risk sentiment is actually very positive across a variety of asset classes, as I mentioned in my prepared remarks, we are still seeing a fair amount of dislocation and opportunity within real estate credit. So that really continues to be an underlying kind of key tenet of our investment focus is being able to redeploy capital with some duration in today's wider spread environment at a lower basis. I think it was going to serve the company well over the long term.

Our next question comes from Steven Delaney with JMP Securities.

I appreciate your opening remarks and Doug especially your macro view. I'll come back to that in a minute. Bob, can you tell us -- I haven't had a chance to look at the Q yet, but on the $129 million loan, I assume that specific reserve is disclosed in the Q. Can you share with us what that might be, the amount?

I cannot. We don't disclose loan-by-loan specific reserves. What we do disclose next quarter the loss, you'll see it flow through the income statement. And Doug touched earlier here on why we don't disclose.

Well, obviously, negotiating with loan buyers is probably a big thing because your tip in you hand as far as what you take, right, if thing -- they just got reserves...

[Indiscernible].

Dumb question. I'll move on to the next thing. The one thing also was positive. To get to be a 5, you have to probably be a 4 first. The full weighted loans dropped almost $400 million, 7 loans now at , it's 11% of the portfolio versus 18% of the portfolio at March 31. Overly simplistic, I know, but we kind of know what happens with 5s, they end up sold or foreclosed or whatever. But if you were to tell an investor, would there be value for our clients and our sales as analysts to pay close attention to migration into the 4 bucket as we go through. We've, obviously, as Doug said, probably got another year or longer in this CRE disruption, but it seems like that's the place. If there's one place we can look to get a picture of the future, that might be. Just curious how you all think about that?

Steve, great observation. A couple of comments, and I think Doug has some commentary here as well. I think you're right, risk ratings are something that we take very seriously. They're an important -- they're not the only tool, but they're an important tool to help us manage our risk book and also convey to the market what our current and future expectations are about loans. . So four is probably the key category. We did have three loans as we disclosed, and migrated from 4 to 5 in the second quarter. We didn't have any inbounds. And if you look historically, and history isn't necessarily a guide of future behavior. But if you look at historically, loans that moved into the 4 category, a fairly large proportion of them actually don't move to 5. And many of them get resolved while being forced. And some actually move back to 3. But it's a really important thing. You're right, folks should pay careful attention to it we do. Doug?

Yes. Look, I think you are hitting on, I think, a little bit of a positive in terms of this quarter, which is, again, we've been saying now for a few quarters in a row that we're really focused on being able to be -- being one of the first kind of through a lot of the challenges. Again, we -- I'll say again. We've avoided to kick the can. We've reduced our office exposure by over $1 billion in the last 15 months, and we continue to have a very clear focus on getting any credit challenges behind us because it just allows us to play offense sooner relative to the broader market. . And I would say, secondly, there was one thing that is worth highlighting in terms of probably the second positive, and that does relate to if you -- absent our credit losses, what our distributable earnings per share were this past quarter. So I think Bob should perhaps just rehighlight some of those numbers just we're really clear because I think that's probably the -- one of the most important, I would say, indicators in terms of the potential earnings power of the company and sort of where we believe to be headed. I think just -- though king of hearing those numbers one more time so that we're sort of all clear collectively on what that looks like.

And so I think the point Doug is directing me to is that distributed earnings before the realized losses for the second quarter were $0.25 a share, which is actually $0.01 higher than our dividend. You were talking -- Doug commented in his general remarks about the general risk environment and the rate environment. And rates are clearly important to us and companies like us. But in terms of our ability to rebuild earnings power, rates are actually a secondary issue. The biggest issue is clearly for us to move quickly through the credit challenge loans, resolving like the best value as possible. We had, at quarter end, $555 million of nonperforming loans. All of those are financed. So we're paying interest on the funding costs and we're not collecting interest. The foregone interest there on an annualized basis, by our estimates, is $0.15, $0.16 a year in a quarter. So that could be a potential upside node. Stephen and asked earlier about the math surrounding CLO cash reinvestment. And then I'd say the third component is we are intentionally carrying high cash levels in order to support our portfolio and be able to work toward the optimal resolutions for each credit challenged loan. But in a more normalized environment, if we succeed in our objectives, we would not be carrying $40 million -- or $310 million of cash. And so if that were deployed, you can make up your own ROE, but if you assume, say, a 9.5%, that's another $0.05 or $0.06 a quarter. So clearly, some drivers. They're all contingent upon executing what Doug articulated and has articulated again and again, and that's working thoughtfully, but quickly through our credit challenge loans to achieve the highest recoveries possible for our shareholders.

Our next question comes from Derek Hewett with Bank of America.

Doug and Bob. I think, Bob, you addressed my -- or at least partially address my question in terms of the sustainability of the dividend, given you're currently carrying excess liquidity and you have additional self-help opportunities from refinancing of certain loans within FL5. But could you just, in general, just talk about your thoughts in terms of how you determine what the dividend should be set at?

Sure. Good question, Derek, and thanks for dialing in this morning. Our view is that dividend level for a company and our company in particular ought to be set at a level that matches its sustainable earnings power. There is also a consideration about distributable earnings, which to say is a pretty close, but not perfect proxy for taxable income and distributing at least 90% of taxable income is a REIT rules based driver dividend levels as well, but let's focus on the first point. We just articulated that NIM currently exceeds our dividend level. Doug described that volatility of distributable earnings is likely to continue as we work to resolve these loans. The team spends extreme effort on that and also in terms of getting our CECL reserve right so that we feel that the net carrying value of our loans are fairly stated. And to Stephen's earlier question in instances where we do resolve a loan that the realized losses closely approximate the CECL reserve. So all of those are variables in the equation, and we use that to formulate recommendations to our Board because at the end of the day, that decision is there's to confirm, but that's how we think about it.

Thank you. There are no further questions at this time. I would like to turn the floor back over to Doug Bouquard for closing comments.

Thank you. Just want to thank everyone for dialing in this morning, and we look forward to updating you in the next quarter. Have a great day.