Starwood Property Trust Inc (STWD) 2019 Q3 法說會逐字稿

Greetings. Welcome to the Starwood Property Trust Third Quarter 2019 Earnings Call. (Operator Instructions) Please note this conference is being recorded. I will now turn the conference over to your host, Zach Tanenbaum, Director of Investor Relations. You may begin.

Thank you, operator. Good morning, and welcome to Starwood Property Trust earnings call.

This morning, the company released its financial results for the quarter ended September 30, 2019, filed its Form 10-Q with the Securities and Exchange Commission and posted its earnings supplement to its website. These documents are available on Investor Relations section of the company's website at www.starwoodpropertytrust.com.

Before the call begins, I would like to remind everyone that certain statements made in the course of this call are not based on historical information and may constitute forward-looking statements. These statements are based on management's current expectations and beliefs and are subject to a number of trends and uncertainties that could cause actual results to differ materially from those described in the forward-looking statements. I refer you to the company's filings made with the SEC for a more detailed discussion of the risks and factors that could cause actual results to differ materially from those expressed or implied in any forward-looking statements made today. The company undertakes no duty to update any forward-looking statements that may be made during the course of this call.

Additionally, certain non-GAAP financial measures will be discussed on this conference call. Our presentation of this information is not intended to be considered in isolation or as a substitute for the financial information presented in accordance with GAAP. Reconciliations of these non-GAAP financial measures to the most comparable measures prepared in accordance with GAAP can be accessed through our filings with the SEC at www.sec.gov.

Joining me on the call today are Barry Sternlicht, the company's Chairman and Chief Executive Officer; Jeff DiModica, the company's President; Rina Paniry, the company's Chief Financial Officer; and Andrew Sossen, the company's Chief Operating Officer.

With that, I'm now going to turn the call over to Rina.

Thank you, Zach, and good morning, everyone. This quarter, once again, demonstrated the strength of our multicylinder platform, with each component of our business, working together to deliver strong earnings performance.

Collectively, we generated core earnings of $153 million or $0.52 per share. Our performance was led by our largest segment, the Commercial and Residential Lending segment, which contributed core earnings of $109 million to the quarter.

On the Commercial Lending side, we originated $1.2 billion of loans with an average loan size of $138 million, all of which were first mortgages.

During the quarter, we received $745 million from loan repayments. These repayments carried fewer prepayment penalties than last quarter, which accounted for much of the decline in interest income this quarter. These cash inflows were offset by fundings of $351 million on new loans and $276 million on pre-existing loan commitments. Our commercial loan portfolio ended the quarter at $7.9 billion with a weighted average LTV of just under 65%.

We continue to finance this portfolio with both on- and off-balance sheet leverage. As a reminder, off-balance sheet leverage, mostly in the form of A notes sold is nonrecourse, match funded and is not cross-collateralized against our retained loan interest. Our Commercial Lending book had off-balance sheet leverage of $3.1 billion at quarter end. If you assumed we financed all of this on balance sheet, the aggregate balance of our loan portfolio would be $11 billion.

On the Residential Lending side, we continued our expansion of this business by purchasing $618 million of non-QM loans and completing our fourth and largest securitization for $546 million. This brought our residential loan portfolio to $1.2 billion at the end of the quarter.

In order to optimize current returns on this business and better utilize our low financing costs, we decided to hold a portion of these loans to maturity. In doing so we classified $479 million of loans as held for investment this quarter. The remaining balance continues to be held for sale and intended for future securitizations.

I will now turn to our Infrastructure Lending segment, which contributed core earnings of $3 million to the quarter. This includes a $2 million loss on extinguishment of debt resulting from the sale and repayment of loans we acquired from GE.

As you know, our strategy has been to sell the lower margin loans in this book and redeploy the capital into higher-yielding loans. As we execute on this strategy, we sold $47 million of the GE loans this quarter, and have $164 million left to sell.

We also received repayments of $237 million, bringing the balance of the acquired portfolio to $1.1 billion from $2 billion. The remaining $446 million of infrastructure loans on our balance sheet were acquired post acquisition. As we work to deploy capital into this segment, we have increased our borrowing capacity into longer-term facilities with lower borrowing costs.

In July, we closed a new $500 million facility with a coupon of L 200, a 5-year revolving term and a term out that extends to the 8-year life of the facility. Subsequent to quarter end, we completed another $500 million 5-year debt facility at our lowest coupon to date of L 175.

Next, I will discuss our Property segment, which contributed core earnings of $29 million to the quarter. All of the wholly-owned assets in this segment continue to perform well, with blended cash yields of 13% this quarter, and weighted average occupancy remaining steady at 97%. We expect our yields to increase next quarter as a result of accretive financings that we completed after quarter end.

In October, we refinanced our medical office portfolio with 5-year floating rate debt. Using proceeds from the unwind of our hedge on the existing debt, we swapped the interest to a fixed rate of 3.3%. We also plan to close today on 6-year supplemental financing for one of our multifamily portfolios. In total, we took out $180 million in gross debt proceeds.

Pro forma for these financings, the wholly-owned assets in this segment are financed with debt containing an average remaining duration of 8 years at a weighted average fixed rate of 3.4%. As of quarter end, these properties, along with those in our investing and servicing segment, carried accumulated depreciation of $366 million or $1.30 per share.

As we have said in the past, we believe these assets have appreciated meaningfully since we acquired them, and the appreciation is not reflected in our GAAP book value. At a minimum, adding back $366 million to our GAAP book value would arrive at our purchase price for these assets. The gains that we believe exist in this portfolio would be an incremental increase to undepreciated book value.

I will now turn to our Investing and Servicing segment, which contributed core earnings of $64 million to the quarter. This business continues to use its various cylinders to produce an attractive return.

In our conduit, we securitized or priced $479 million of loans in 3 transactions this quarter. In our servicer, we continue to add CMBS 2.0 and 3.0 deals to our named portfolio. As of quarter end, we were named servicer on 180 trusts with an unpaid principal balance of $89 billion of which 95% now represents 2.0 and 3.0 deals.

And finally, on the segment's property portfolio. We continue to harvest gains as these assets reach stabilization. During the quarter, we sold 1 property with a cost basis of $38 million for a core gain of $9 million, which is net of a $4 million noncontrolling interest. As a reminder, our GAAP gain on these assets differs from our core gain primarily due to accumulated depreciation, which we include in our GAAP to core reconciliation.

This portfolio ended the quarter with an undepreciated balance of $337 million across 20 investments.

And now turning to our capitalization and dividend. During the quarter, we made significant enhancements to the right side of our balance sheet, further reducing our reliance on repo lines and exposure to margin call risk.

In August, we issued our first CRE CLO, which provided $936 million of nonrecourse term match financing for 21 of our existing loans totaling $1.1 billion. The structure has a day 1 spread of 134 basis points over LIBOR, a day 1 advance rate of 85.1% and a reinvestment feature, which allows us to maintain this advance rate for at least 2 years.

In July, we completed our $400 million 7-year Term Loan B, which priced at L 250. This loan replaced our $300 million Term Loan A, which we repaid this quarter. The transaction increased our liquidity, extended the tenor of our financing and unencumbered $486 million of our assets.

Concurrent with the term loan, we entered into a $100 million 5-year revolving credit facility to replace our current revolver. We ended the quarter with a record $8.5 billion of undrawn debt capacity and an adjusted debt to undepreciated equity ratio of 1.9x.

As for our dividend. For the fourth quarter of 2019, we have declared a $0.48 per share dividend, which will be paid on January 15 to shareholders of record on December 31. This represents a 7.9% annualized dividend yield on yesterday's closing share price of $24.21.

And finally, I would like to conclude my remarks with a few comments about the new accounting standard on current expected credit losses, or CECL. As you have probably heard, CECL replaces today's incurred loss model with an expected loss model for determining the allowance for loan loss under GAAP. The standard is effective for us on January 1 and applies to both our CRE loans as well as our infrastructure loans. It does not apply to any of the loans or securities that we report at fair value.

For our CRE and infrastructure loans, we will not only have to record an allowance on the current funded balance of these loans but we will also have to record a reserve against our unfunded commitments. The adjustment will go against equity at the date of implementation and will thereafter be reflected in GAAP earnings. Consistent with our current policy, core earnings will not include the allowance. We will provide you with a more detailed update on the status of our implementation efforts next quarter.

With that, I'll turn the call over to Jeff for his comments.

Thanks, Rina. CRE fundamentals and loan demand remain strong. Morgan Stanley wrote last week that private real estate funds have $334 billion of dry powder compared to $205 billion in 2014. And that institutional allocations to real estate continue to rise. Transaction volumes have remained elevated the last 5 years. And after a strong September, the United States is the only G10 country where volumes are up, albeit slightly year-over-year, led by major markets, which is where the bulk of our loan portfolio has always been located.

Dry powder and transaction volume, combined with lower interest rates, has led to significant loan origination activity. And with improved cap rates and property prices increasing 6.7% year-over-year, our existing property and loan portfolios continue to perform very well.

With that robust backdrop, as Rina alluded to, our internal valuation of our own property portfolio is up over $120 million from last quarter, and now has a fair value in excess of $800 million over our undepreciated carrying value. More than half of which is in our 15,000-unit Florida multifamily portfolio, which has performed far in excess of our underwriting assumptions.

Strengthening CRE markets have made it difficult to add core stabilized properties with double-digit cash yields to our Property Segment. Our $3 billion of wholly-owned properties are earning a current yield of 13% today. And as Rina mentioned, this will increase significantly as a result of the successful refinancing of our medical office portfolio last week and the pending upsize of our debt on our high-performing Florida multifamily today.

We will continue to model the earnings power of holding these gains across our property portfolio versus realizing them and reinvesting incremental proceeds. And keep you up-to-date on any changes in our plan going forward.

We deployed $2.5 billion in capital this quarter led by another strong performance in our core CRE lending business, which deployed $1.2 billion, and we expect a significantly stronger fourth quarter. We continue to see great opportunities internationally, where we wrote 3 new loans for $554 million in the quarter, with a 7% unlevered return and a levered return in line with our current yield targets. Our portfolio continues to benefit from LIBOR floors on loans originated when LIBOR was well above 2%. 84% of our loans have LIBOR floors. Our weighted average floor is 136 basis points and 25% of our floors are in the money today.

As you'll see in the supplemental deck that is posted to our website, the counterintuitive result of our existing LIBOR floors is that our earnings will increase in our primary -- primarily floating rate loan book with either an increase or a decrease in LIBOR. And we will actually earn more in a decrease in LIBOR than an increase.

Our Commercial Lending portfolio continues to perform very well also with no downgrades in the quarter, and a slight improvement in our overall portfolio risk rating to 2.62 on a 5-point scale.

With all the WeWork headlines in the last few months, I wanted to take a moment to talk about the loan we made on 424 Fifth Avenue, the old Lord & Taylor site, which comprises 95% of the WeWork exposure in our Lending portfolio. We own a $229 million of the $500 million first mortgage loan, and we own the entire $150 million senior mezzanine loan. There's a $250 million junior mezzanine loan, subordinated to us as well as $387 million of equity. We are very comfortable with our last dollar fully funded loan basis of 51% loan to cost on what will be a finished trophy Midtown building with a 15-year corporate guarantee from WeWork.

As Rina said, our non-QM Residential Lending business acquired $618 million of loans during the quarter and completed its fourth and largest securitization to date, locking in the lowest cost of funds we've realized to date. This high-quality portfolio has an average FICO of 720 and 65% LTV. We expect to continue to increase the scale of this business in the coming quarters.

Finally, only 3.1% of our loan portfolio is to dedicated retail assets and our portion of senior loan in the American Dream represents more than half of this exposure. Most of you have read that the theme park opened in October and the retail, which has exceeded our underwritten leasing estimates, is scheduled to open in March.

Based on our conservative leverage position in the capital stack at 36% loan-to-cost, the continued leasing momentum of the project and the pledge of interest in other assets of the borrower, we continue to feel very comfortable with this loan.

Our Energy Infrastructure Lending business, SIF, started the year slowly and cautiously, following the equity selloff and spread widening in December. We spent the bulk of the first half of the year focused on selling down the lower-yielding assets we acquired in the GE purchase and setting up financing lines. As Rina mentioned, we have closed 2 new $500 million, 5-year financing facilities at sequentially tighter spreads, giving us ample financing to execute our business plan into 2020.

With these financings in place, we have a robust pipeline of well over $500 million in loans for Q4 at a blended IRR that is very accretive to our overall return threshold.

As a result of the $1.1 billion CRE CLO we talked about last quarter, which priced at the tightest financing levels and best structure of any deal done post-crisis as well as A Note sales we have made this year, almost 60% of the financing for our commercial real estate lending segment is now off-balance sheet.

Our on-balance sheet leverage dropped during the quarter as a result of this, back to just 1.9x, and our leverage, including off-balance sheet debt is still a modest 2.9x today despite having high quality, higher leverage businesses like non-QM and energy infrastructure lending on our balance sheet.

We will continue to conservatively lever our business, looking for lending and equity investments that do not require outside financial leverage to meet our return threshold. Reducing our on-balance sheet debt leaves us today with a record $8.3 billion of financing line capacity across our businesses and the ability to scour the market for the most optimal form of financing for every asset without pushing on leverage or structure.

Finally, we've looked at investing in investment-grade SASB CMBS, single-borrower CMBS, and in CLOs, but have chosen not to allocate capital to this strategy. Rates are low, spreads are very tight -- spreads are at very tight historic levels, bid offer spreads can vacillate and it requires multiple turns of spread mark repo leverage to achieve returns significantly above what we get by risklessly paying down our bank lines. We instead benefit from having multiple businesses, allowing us to invest our available capital into loans and assets we source and underwrite ourselves.

With that, I will turn it to Barry.

Thanks, Jeff. Thanks, Rina. Good morning, everyone. I can make my comments short. I say that and I always talk too long. But this morning, I'm going to do that.

The overall environment is interesting out there. The property markets continue to have great strength, both in the United States and in Europe, particularly the office, industrial and multifamily markets, where we've seen accelerating rent growth across our own portfolio in our equity books outside of the REIT.

The 2 asset classes that remain somewhat challenged to our retail and the hotel space. Hotels, primarily because of overbuilding more than a lack of demand. And the supply in multifamily actually remains fairly handleable or benign. I wouldn't say it's totally benign, but it's -- there's more net absorption in every market than there is construction, save the Northeast. There are pockets of problems, New York City, high-end retail -- residential, as you know, the West Coast, San Francisco. And we are worried, increasingly nervous about the impact of a Democratic victory and raising rates on tax rates on the wealthy, in cities like San Francisco and like New York, which would cause an exodus of wealth and of relying -- ultimately, the burden of carrying the social services of these blue states will fall on an ever shrinking population.

It will be a negative cycle, reinforcing cycle. It's beginning to impact our lending thoughts because will you really want to be long expensive building at a low cap rate in New York City if tenants like JPMorgan decide to hop -- skip town. And you're seeing that reflected in the lack of strength of their housing markets in and around New York City, in particular, and in all the blue states with the removal of the deductions of taxes, but I think we can all assume that with a Democratic victory, there's no chance taxes aren't going up on the rich in some capacity.

So what you're seeing also is tremendous volatility in interest rates, much more so in the last couple of years. 2 days ago, the 10-year was 160 something, and this morning, it's closing in on 2. It doesn't really impact us at all, but it's interesting to see the uncertainty in the markets, the melt up of the stock market. And we were just concluding our LP conferences on our equity funds, and I referenced the Beyond Meat chart where stock went up 820% off its IPO. And now, it's down by 2/3 from its IPO. Just because the market's rising, don't mean they have to stay in that position. Clearly, the equity markets are ignoring the prospects of a far-left victory at the moment. And at some point in the next 12 months, they may reassess that depending on what happens in the political arena.

So staying closer to us, I mean we're very comfortable that our loans -- our LTVs will probably drop on assets that we're lending to because of the rise in rents across pretty much the country, in most major office markets.

Also if the economy weakens, rates will stay low. And I think they're driven right now more off of European condition than they are off the U.S. condition. And in Europe, Germany is probably focused on some fiscal stimulus to raise their -- pull them out of a manufacturing slowdown. And they've been quite the laggards when you look at the competitive landscape of what governments have done across the country, what the Japanese have done, what the Chinese have done, what the Americans have done.

Clearly, the German -- or the ECB has done far less in fiscal stimulus than the rest of the world. And their economy is kind of showing the impact of that double-wallop, that and also the Chinese trade negotiations and their dependence of -- on foreign trade, which is so much different than our economy here. But I think that means that if Germany were to get going, it probably pulls U.S. rates higher even if the U.S. economy slows since I believe we're funding our -- we're pricing off of the bund.

So coming back to us, we've never been stronger. And we have, as Jeff mentioned, $8.5 billion of undrawn facilities, ample cash on our balance sheet, our multiple credit business lines are functioning really well. Really excited about the growth in the resi book and the quality of what we're doing, the returns that we're getting. And we are moving to hold more of that paper than we had in the past. We're going to increase our on-balance sheet rather than securitizing all of it because the yields are in excess of what we can return in our Commercial Lending segment. And the risk, we believe, is at least as good or maybe better.

The energy Infrastructure business has actually cost us probably in excess of $0.10 and growing this year in the year-over-year earnings. Primarily because, as Jeff mentioned, we shut down originations. We have a facility for -- a 2-year facility when we bought the company from GE, but the loans are 5-year loans, and we don't mismatch maturities. So we basically said we're going to wait. So the team put together these 2 financing lines that are of longer duration, now we can comfortably make these loans, and we're off to the races where the book has never been bigger. We'll hit or exceed, I expect, our origination goals in that business and it will add accretively to our earnings next year and going forward. So pretty excited about having a broken wing business contribute and start flapping its wing and helping us go in the right direction.

The property book is beyond belief good, particularly our multifamily assets. So they are -- we don't know what to do with them. I mean there's -- we have this more than $500 million gain. And cash and cash yields that are 14%, 15% and only going up.

So we could harvest them. We don't know how we'd replace them. And it would require -- if there was, let's say, $800 million of embedded gain -- or capital and gain coming back to us, from the sale of those assets, at our leverage ratios, that's another $4 billion, $5 billion of that loans we have to make. And the average duration of our loans is short. These assets will live forever. So we'd have to put $4 billion at every 2 years or 3 years.

It's a high burden on top of the 6 or so 5 or 6 we'll originate this year. And we did notice, by the way, and Jeff mentioned -- Rina mentioned it, and it's something I actually just caught on to, is that our largest competitor does everything on balance sheet, and we do these A Notes. And so our volumes are understated because if we actually did finance the company the way they finance their company, our loan book would look at $11 billion, not the $8 billion -- $7.9 billion that you see. And the number of originations might look different, too. So it's really -- it's just the way we presented ourselves, and we're going to correct that going forward. Because it's really an $11 billion book.

The thing that we also don't do in scale, we don't finance ourselves and there's a material portion of our competitor's book is loans to themselves even though they're quite large. It's not a bad thing. I'm not saying it's a bad thing, it's just a different philosophy. And we've, on occasion, done it in very small numbers, way back when, probably 7 years ago. There was a 13% mezz on -- I think it was a $250 million position. The $200 million to $250 million of -- on Mammoth Mountain, which we owned. We sold a piece of it off to Apollo and took the 49% loan. That asset sold for $810 million. Our last dollar exposure was $250 million. So when there's just stupid paper available, we have occasionally done it, but in very small scale. Certainly, 1/100th of the size of what our competitor's done.

And I think the other interesting thing in the book is the $90 billion servicing portfolio, all new. And I will mention B pieces because some of our peers -- we are the biggest player in servicing. So we should be the biggest player in B pieces, right? We're not. We don't really think it's a tremendous business. It's insurance that's mispriced. Not bad, works in the bull market, but it's not going to be ever a core business for us. And so will we always participate, I think we're 5th or 6th in B piece purchases, Jeff? Something like that. And we cherry-pick the ones we want to buy, particularly the ones where we are a contributor to the pool, and pretty much move on from that.

So I don't have any other comments, I think, the company is doing quite well. We've completed a 3-year plan. We're quite sanguine on our future. We are looking at things to buy and have tried and failed. But perseverance is genius in disguise. So there'll be other things we'll be looking at in the future. So with that, I'm going to stop. Thank you.

(Operator Instructions) The first question is from Jade Rahmani of KBW.

This is Ryan Tomasello on for Jade. Jeff, we appreciate the comments in your prepared remarks on 424 Fifth Avenue. I was wondering if you can say what you think ultimately plays out with that asset with WeWork walking away from the lease and some individuals believing that the sponsor group overpaid for the asset by several hundred million. The 51% loan-to-cost metric you cited, can you state what value that's based on? And if it includes WeWork's purchase price plus any anticipated renovations for that asset?

Before you -- one thing, they can't walk away from the asset. There's a 15-year corporate guarantee at this point, the sponsor is long, I think, will be long something like $19 billion invested in this company. So it is their single largest credit liability, as a single liability now on their balance sheet. But they can't walk away from it. And they can sublet it or sell the building. Obviously, they don't own the building, there's a different group that owns the building. And that has got institutional capital to the tune of what, $380 million or something?

$387 million.

They've spoken to us, and they believe that there's equity like they don't believe they've actually lost money. And they think there's tremendous demand for this space. So -- and we've been approached to buy our note, and we're noodling on whether we want to do that. It's actually a pretty good note. So Jeff can give you the stats.

Sure. I laid out part of the cap stack earlier, we're in the $500 million first mortgage. We are the $150 million senior mezz. That was $650 million, last dollar. The total cap stack includes, as I said before, $250 million of junior mezz and $387 million of equity for a cap stack of $1.287 billion, and that's where I get the 51% of cost, using $650 million out of $1.287 billion.

Okay. And I guess just moving on to the lending markets overall. What's your current view of the levels of competition and underwriting standards? Do you think that nonbank lenders are increasingly getting more aggressive?

And then regarding 4Q, you mentioned you're expecting a strong investment quarter. Can you provide us with any parameters around what you're expecting for origination volumes as well as repayments in the fourth quarter?

Sure. I would say, first of all, we haven't changed our standards in the 10 years since we started the company and the 5.5 that I've been here. We have a very thorough and difficult investment committee process through 2 investment committees. We use tremendous amounts of 30 years of data and information from Starwood Capital Group and Starwood Property Trust. And our standards are what they are. The good news is that the volumes are fairly high, and we are seeing a decent amount of loan activity for everybody to choose from, and we're all going to pick and choose where we want to be within that.

The nonbanks -- people think that they're being more aggressive. I do think we're getting pushed to tighter spreads today. Fortunately, we have tighter liabilities, so we're able to earn similar spreads. But one of the things you have to realize is not everybody has the same cost of capital. There are a number of nonbanks with separate accounts with a lower return hurdle than we have, and they're going to look at slightly tighter deals than we will look at. And those are deals that may cross over into our space.

So there are times where we will feel like spreads are a little bit tighter than they would have been. But oftentimes, it's not because -- I don't believe it's because people with the same cost of capital as us are loosening their standards to go to tighter spreads, I believe that it's because there's just simply a number of different costs of capital from banks at the lowest cost of capital to some separate accounts at a more moderate cost of capital and then our cost of capital being a little bit higher.

So it's hard to say that standards are getting worse and that loans are deteriorating when everybody in our space seems to write more every year. We'll be within 10% of what we wrote last year. Last year, every one of our competitors did the most amount of loans that they've ever done. Loan volume is high. There will be over $500 billion of transactions again this year for the fifth straight year, only time since 2007. And so there's plenty to go around. Everybody is going to choose their buckets. We are going to focus on credit more than spread, and we're going to find the credits that make the most sense for us and our cost of capital.

The fourth quarter, we never give an exact number. I would say, we think that we'll do at least somewhere in the neighborhood of 50% more than what we did this quarter. Fourth quarter has always been seasonally a stronger quarter for us, and we believe, without moving our standards that, that will be the case.

Europe picked up in the third quarter fairly significantly. It will be a little bit smaller next quarter, but we're hoping in the first quarter, that will pick back up. We have people in Australia, as we've talked about before now, and we're hoping to continue to grow the international book. So the number of things that we have to look at is still robust. We hired more originators this year to be able to get through it. And I think that somewhere in the sort of $5 billion to $7 billion annual run rate is what I would expect us to be able to continue to achieve without lowering our standards.

And then just one, if I could. Barry, you mentioned that you're looking at things to buy, I think you specifically said. What types of things are you looking at? Should we expect it's more infill to the current business focus, in terms of property...

I need some furniture for my new apartment.

I think the GE deal -- I mean the infrastructure deal from GE was interesting. Is it stuff like that where you're expanding the actual business line focus or more infill?

I would say it's more infill or REIT-able business lines, not everything we look at middle market lending, some of these other stuff, we can't actually fit in our structure. We can fit it in our structure, we have to go in the TRS. But we may not be the best buyer for it because of our tax hit we pay. So it requires some significant restructuring.

It's interesting. There are a lot of players in our space. And I'm not sure we're all needed. So we also -- the restructuring of the federal agencies might present interesting opportunities to us. We are -- I think it's known that we are looking at purchasing an originator in the resi space, so we can be a fully-integrated company in that space, and we're just -- something is under contract and we have waiting for regulatory approval. So building out these businesses so that potentially we could do another Starwood Waypoint Homes, Invitation Homes spin from ourselves, it's something we would look at.

With our resi book, if it grew large enough to be a standalone company, that the reason -- there's only one reason to spin it. It just runs at higher leverage levels than our business. And the screens of our company make us look odd even though I don't believe we're taking on any excess leverage. Being on the Board of Invitation Homes since we spun it out of us, that company, which has done quite well, had never thought it could trade where it's trading because they thought the market would puke on the leverage levels of the company. And we had big fights at the Board meeting about should they use the money to grow, would there -- any free cash flow, should they use the money to pay down debt, should they use the money to increase the dividend, which is de minimis and well below the yield on other equity REITs. And I argued that you can't possibly make a dent in the leverage levels and it's really the ROE of the business that is super attractive, and there's no excess risk being taken by having those leverage levels. And so the company's stock has performed great this year as the company now has passed its NAV.

And so the point is that the market said, we understand a high leverage structure when you're not taking a lot of risk and drove the market value of the equity up, so the average fair value of the debt versus equity market value of the equity is lower. So I'm saying that maybe the market would allow us to have a giant resi book inside of ourselves. Don't know -- but we want to keep the opportunity of maybe spinning it out, that would require us to be, as I mentioned, a fully integrated group, which would require us to be in the origination business as well.

So we're working on that, and we've got a great team. They're running a great business. It's adding -- it's accretive to our platform, and the IRRs are higher than GAAP allows us to tell you they are.

So which is funny because they really are very interesting IRRs and GAAP doesn't allow you to forecast refinancings and other things you anticipate doing as the corpus of the bond -- the bond portfolios get paid off. So it's a very good ROE business for us.

Ryan, I wanted to revisit yourself position on the equity paid too much for 424 Fifth Avenue when we were talking about WeWork before. If the tenant were no longer a going concern, and we had to relet the office -- if we relet it $30 lower on our blended rents, we would still be south of 70% loan-to-value and north of a 7% debt yield on that building. So we have supreme confidence that our loan is better than money good. And by the way, we have a LIBOR floor that's 70 basis points and the money on that loan as well so...

Again, we're not the owner. So we don't get to set that WeWork rent. But again, they lowered that rent. Don't forget, WeWork assumed they were going to occupy most of that space and then do enterprise deals on the balance of it. So the enterprise tenant wouldn't be taking $30 less than WeWork was paying, they'd probably be paying $50 more than the space was going to be built out for. The money going into the -- that we're funding and wasn't drawn at closing was done for the fit-out of the building, which was to do enterprise for the non-WeWork headquarter space. So you're going to have a very nice building, and that's probably, what Jeff saying, it's probably a 50% drop of what they expected to rent those spaces for -- to the Amazons and other people that were interested in being in that space with them.

So we feel very comfortable.

The next question is from Steve Delaney of JMP Securities.

Regarding your comments about the reinvestment challenge on real estate gain realizations. I guess you guys have -- Barry has been doing this a long, long time, and you can either buy it or build it. And there have been some recent news articles about a property you own, called The Mall at Wellington Green about the potential for a major redevelopment. I wondered if it's possible you could comment about that particular property, but just kind of in general, the returns that you might see on especially urban infill and redevelopment as a deployment of capital and what those returns might look like.

Thanks for your question. We're not in the development business. We're a lender. We happen to have an asset there in Wellington that the Nordstrom store left and allowed us to reposition the mall -- potentially reposition the mall. In that case, we proposed a lagoon, a crystal lagoon and building a beach around that as an amenity to drive traffic through the mall. The town is very excited about, the mayor. But we have to look at the incremental capital and how much it's going to cost us to do and what we think the investment returns might look like and what the shape of the -- probably the outcomes might be.

I mean in retail today, that's complicated. There are so few comparables for sales of malls that if you think you're going to sell this property once stabilized at a 7 or a 6 or a 10, 11, 12, you either go forward or not go forward. And there's just not a lot of clarity about that situation. So we're -- it's something we're evaluating in the context of all of our assets. We look at our redevelopment potential but mostly, this entity doesn't do that. I mean this is not what we anticipated doing in this vehicle, that's more -- development deals are really done in our equity funds. So they're not done here.

Okay. Jeff, switching over to the resi side, obviously great progress there. Bloomberg had an article on Monday about -- and said, we're pushing up to $20 billion year-to-date. What inning are we in? And I guess in specifically in QM. Do you have a sense for what that annual origination securitization market could be at?

Yes, I think it's a fraction of what it could be. You have a non-QM patch happening at the agency that if I'm right, it's a multiple of almost 10x that of potential available capital that come back. I think you're on $18 billion to $20 billion, up from $12 billion or so last year, year-to-date. There's a handful of large private equity-type sponsors like us who are investing in the space. Ultimately, I think it's potentially a couple hundred billion-dollar business. And I think that the business could certainly grow and a lot of it depends on what the agencies decide to do in terms of how large of a portion they want to be.

One of the things that pushes people to non-QM is if they fail the 43% debt-to-income test. And typically, with a lot of self-employed people in the economy, you're going to have a lot of people who don't state the full income because they have expenses that they run through their business or whatever else that is. But who are good borrowers and can put down significantly more money. These borrowers are putting down 35% on their houses versus agency borrowers, who are putting down 10% to 20%. These borrowers have higher FICO scores generally, but they fail some portion of the test that would put them into the agency bucket.

So assuming that, we will still have a lot of self-employed people in this country and demand for real estate loans with a slightly higher coupon on the agency side, it could grow a little bit, but what's really going to drive volume for someone like us is where the agencies decide to play, how much they open their box or close their box and what that leaves for the rest of us. So today, it's a kind of cute business, where we'll do $2.5 billion or so a year, and I would love to do double that, but it's hand-to-hand combat every day, trying to get close to a bunch of originators to be able to do that smartly.

Yes. So shifting from the consumer side, where obviously the regulatory and GSE issues, have you guys looked at the SFR product, the rental, the single-family rental product? Any idea of whether that's an opportunity and how those returns might compare to in QM, which is obviously more competitive right now?

Yes. Yes. Barry, knows that space better than anybody, who -- we obviously spun out Starwood Waypoint Homes...

Exactly.

And Invitation homes, and they do a great job of that. They manage it extremely well.

Yes. And I'm thinking about the debt, I'm thinking about debt, not owning homes, by the way. Just the first mortgages.

Yes. I've got you. We have not spent a lot of time there. A lot of our competitors are looking at things like that and fix and flip and some other businesses. We've sort of stuck to our knitting today. We really like the demographic of a 720 FICO, 65 LTV rather than the professional homebuyer, not that there's anything bad about that. We just are extremely comfortable with the credit, and we know what these borrowers did in the last cycle when things went south, and these borrowers tended to perform very well, best-in-class in terms of how they got through the crisis, and we're sort of leaning on that if we ever headed back into a crisis. We want to make sure that we're in what we believe is the highest credit quality part of the Residential Lending business.

And Barry mentioned that you, and Rina did, too, you put some loans on as held for investment, held to maturity. Just to close out, the FHLB, I mean there's obviously been some positive suggestions in the Treasury's report and everything else. I mean it seems to be that access going forward long-term is huge for your ability to put whole loans on your books. And with that, I'll drop off. But just wanted to get your final thoughts?

I would say that it's beneficial, but not required. And we've lined up another facility to replace them potentially if that does go away. I would say the odds of it going away are less than they were before. But obviously, this is somewhat political, although this can be done with -- what they call it, administratively, it's not a -- has nothing to do with -- doesn't go to Congress. It can be just done by the administration. And we obviously it's never -- it's unprecedented to actually eliminate a business unit that's not costing the government anything, and there's no default.

So it would be a first, if they turn them off. But regardless, we've modeled it going away. And we're modeling even the loans held-to-maturity and the sizing of that book based on them going away. So we think it's the wrong move for the government. We think they shouldn't do it, but they do it, we'll survive. And the IRR -- the ROEs will be a little lower. So it won't be tragic, probably will not grow to be the business we expected it to be, at least on the balance sheet. The securitization business will stay around. They just have a different ROE.

It's good to keep private capital extremely involved. They've done a great job on the bottom of agency mortgage securitizations by selling off to private capital and keeping private capital involved in sort of first loss, and the agency business, I think it would make a heck of a lot of sense to keep private capital involved in competing on the non-agency business as well. And if you start taking away things like that, it makes it more difficult, but we can still compete in the securitization market. We have multiple lines, as Barry said, where we can finance these loans. So we're very comfortable with that position.

The next question is from Stephen Laws of Raymond James.

A couple of things have been touched on, and I apologize if I missed this, but I noticed you refinanced some of the real estate assets. More optimized your capital and you're going to redeploy that. Can we talk about how much more opportunity there is to do that, given the appreciation in the real estate book and how we should think about that capital being deployed or this taking place over the next 12 to 24 months?

The good news is we have long-term fixed rate debt on a number of these properties. And until that rolls, it's a little bit more difficult to do that. The medical office portfolio, we're able to do a CMBS financing and upsize that and take out a significant amount of money. We were able to add on to our first multifamily portfolio. We believe, at some point in the future, not too long, we'll be able to add on to the second multifamily, but the master lease portfolio has long-term debt. And we just refinanced our Irish portfolio last year at 1.9%, which was an incredible achievement for a fixed rate of 7 years to get that kind of rate. So when the debt is rolling, we'll be able to take a little bit of cash out and redeploy that when we redeploy that obviously it's sort of free equity as we put it back out, and then it will be accretive to earnings, but there's probably not a lot on the front burner in terms of refinance potential in the portfolio.

Okay. I appreciate the commentary there. A follow-up, I know you provided a number of details on energy infrastructure earlier, but I saw you sold 1 asset or some sales in the quarter. Can you talk about where you are in kind of better positioning the original portfolio and cycling out of some of the lower-yielding assets that came in and kind of what you have left to accomplish on that front?

Yes, sure. We have got rid of the bulk of the lower-yielding assets, there's very little that we would want to sell if the bid was there. And as Barry said, getting the 2 lines that we got has really opened the gate for us to start to add accretively to this portfolio. I have a sheet in front of me that is well over $500 million in loans that we expect to close in the fourth quarter. And if we ran at a rate like that, we underwrote this thinking it was $1.25 billion to $1.5 billion a year. So $500-plus million in a quarter would show you that we think there's a pretty good opportunity here. And the levered IRRs are above what we originally underwrote. So we're optimistic. Things can change quickly. But right now, the -- what's in front of us feels really good in that business today.

The next question is from Rick Shane of JPMorgan.

Look, it's year-end 2019, we're at arguably a little bit of a crossroads in terms of outlook and economy. As you guys are planning your 2020, and I suspect you're deep in the middle of that right now, curious how you think about the different scenarios and the different paths you can take.

Actually, from a lending perspective, our businesses, I am fairly sanguine. I think it's much more concerning on the equity investment side than it is on the debt side. I think low rates have actually -- the first half of the year, our transaction volumes were down like 10%. I think around the world, capital flows have been adjusted or rejiggled because things like the Chinese aren't here. The Aramco IPO probably if it takes place in any scale will mean that there'll be greater participation in the U.S. markets by Middle Eastern investors than we've seen, but I think we take the general view that the economy is going to slow and fairly significantly next year.

Not go negative, just slow. Slow because of the election primarily and the difficulty in determining what's going to happen. And it's a toss-up. We don't even know who the candidate -- I'll say, candidates on either side are going to be, given the impeachment process. So I think that specter means -- you see this unbelievable thing going on all over the world actually, but it's worse in the U.S. The consumer confidence remains relatively high. CEO confidence has been not as low since 2009. And so they're looking at a world with all of these problems and all the rants of the left, and the far left, and they're worried. And so worried means you wait. And the portion of the GDP that's not health care, investment spending will slow. CEOs will wait. They'll say, okay, let's just wait and we'll see what happens with the election, and then we'll plan for the next 4 years.

So I don't see any way the economy won't slow. It's not in historical data. I mean people look at this -- the pundits come up with "Oh, this is going..." It's ridiculous. I mean there's never been an election this polarizing since I've been alive, because 2 ideologies are going head-to-head in a -- 2 extremes. And people keep forgetting 43% of the country is independent. It's the largest party, and there is no party for the middle of the country. And they're telling Bloomberg, you have to run as a Democrat, right? Ross Perot Jr. was a billionaire and probably not the greatest candidate in history, and he got 19% of the electorate right? And the independents at that time were about half the size they are today.

So I mean the country needs somebody in the middle, in my view, sorry, I'm being political on an earnings call. But the country is fiscally conservative and socially liberal, the vast majority of people are in that bucket, especially kids. Kids may not be social -- fiscally conservative. But that -- I think from our perspective, it's not an issue, from the REIT's perspective. And we continue -- I don't see any of our business lines being impacted by the craziness. All we have to do in lending is get our money back. We don't have to make money and think about the increase -- and so having our biggest equity assets being the affordable housing space, where rents can only go up based on incomes of -- in our -- average incomes in towns like Orlando and Tampa. And that we're just feeling really good about that.

So we don't really have anything else that we feel like we're exposed with. We feel the team has done an unbelievable job on the balance sheet of the company. And it's the best in the industry by far.

So we're positioned for safety. We've always been there. And you may give up slightly excess returns. But 10-year track record of this company is 12.6% or 12.8% compounded per annum for 10 years, total return. That includes our spin-off, by the way, 240% total return. It's admirable. And if you levered that, you'd be at -- the best hedge fund in the United States. You'd have a 20-plus return for 10 years running, and you'd be running $60 trillion. So I wish I'd done that trade myself. So I have been long in the stock, obviously. But we're happy.

We're concerned about lack of lending discipline. It's not so bad. I've been through a lot of these. I mean it's not crazy. There's a few outliers here and there. People do some nutty stuff. They lend to -- sometimes we have to lend to people that banks won't lend to. But we'll do it based on our understanding of the asset values. So we lend to Kent Street, right? So in New York, we'll leave it at that. Jeff is shaking his head at me. So but a bank wouldn't make the loan and we made it and we got paid off.

I guess I would answer it differently. Barry, obviously has a lot of thoughts on where the world is going to go, and he's usually pretty spot on. But I think our job as management here is to make sure that we perform well in any environment. We now will perform well if interest rates decline or fall and where we make the least money is if they stay here. If credit spreads widen like they did in December of last year, it's a great opportunity for us to add to our book.

We noted earlier, we have $8.3 billion of capacity to add today. And that's a -- that gives us significant firepower. We'd love to add more property, more into our Property portfolio, if cap rates were to widen, that would be a wonderful opportunity for us. And if and when spreads tighten, we will probably take some gains like we have in the past and try to figure out how to redeploy them. So I think as we look to 2020, we've significantly reduced our exposure to credit marks by the moves we've made at the CLO and the A Note sales, et cetera. And we're really positioned extremely well for either a rise and fall in rates or regardless of what happens in credit spreads. So we're pretty excited about the positioning of the company going into the new year.

Our final question is from Tim Hayes of B. Riley FBR.

As a D.C. resident, big fan of the cover of the supplement. And your largest loan this quarter was a $300 million construction loan for The Wharf project. And your second largest loan was also a construction loan. Can you just touch on how you think about construction lending at this point in the cycle? What percentage of your book consists of construction loans and what the LTVs are here? And any other color on structure?

Sure. There's a lot of schools of thought on lending. And when we're in construction loans, we are typically in a much lower loan-to-cost than we are a loan-to-value on a regular loan. So if the risk is completion when we come out on the back end, we certainly feel very comfortable with it. That said, our Board of Directors has always been very concerned about future funding risk.

One of the things that we have going for us, having this very large diversified book is we have money coming in from lots of different segments of our business. So we like to think that we don't want to ever have more money going out in any quarter in future funding than we have twice as much money expected to come in at that time.

As a percentage of our overall loan book and as a percentage of our assets, we have a significantly smaller construction book than what we had back in 2013 and 2014. And we have -- we never give the exact number, but I would say we're sort of right in the middle to slightly above the middle of our historic average in terms of construction exposure. Our future funding is relatively small versus the amount of money we have coming back in our $17 billion balance sheet. So we feel very comfortable with that. We have a band from our Board, and we're 25% below the top of that band, and we've been at this position or very close to this position for most of the last couple of years.

So I don't think it's changed much at all. Our book has got bigger obviously which allows us to do a little bit more construction. Our overall balance sheet is now close to $17 billion. We think about it a lot. We're very comfortable with the risk in terms of the LTC versus the LTVs on other opportunities. Then we're extremely comfortable with the future funding exposure today.

One of our directors was saying he was hoping these guys would default. Then we'd get to buy brand-new buildings at 2/3 of what they cost to build. But that's not going to happen. But we are very much, as you know, we founded iStar. We changed its name to -- it was actually Star Financial, it became iStar. And then they bought a bank called -- I forget -- what it was called? Anyway, and they blew that deal because what happened was, while they got a discount on the $3 billion of funded loans, they had to fund all those draws at par. And that wouldn't have been a problem except Fremont. It was called Fremont. They -- when the world ended, they didn't have the sources to fund those -- make those commitments to the -- on the construction loan. So the repayments stopped. We know the drill. We saw it go bad. And so we're pretty careful about our exposure to the construction market despite the fact that we like our projects, and we like our attachment point. You mentioned The Wharf, I mean wouldn't you like to buy it at 2/3 of what it cost to build?

Absolutely.

So it's not going to happen. But we're going to be okay with it.

We typically do 10 or 12 loans a quarter, and if looking back in the last year, a year ago, this quarter, we didn't do any construction loans. In the first quarter of '19, it was about 10% of our book. Next quarter, it will be a lot smaller percentage than it was this quarter. So 1 or 2 construction loans can make the number look like it's a bigger quarter, but it's not a significant trend.

Got it. That's all really helpful color. And then just 1 quick follow-on, just about the Dublin portfolio sale. Just curious if you still intend to sell that portfolio before year-end? And if there's any changes in expectations around the type of gain you might receive there?

It's actually a topic for Monday. We have bids and we're going to decide what -- for us, like, it's like, okay, what do we do with the money? So we're flush with cash and capacity and I wish we could double our pace of originations that met our criteria, and we're going to try. But we have a nut we have to cover in terms of an ROE, and we don't want to lessen our credit underwriting standards. So we're going to try. And one place we look to pick up volumes and the team is pretty bullish on, is offshore.

And we have an office in Australia. We have a team, I think, 8 people or something like that now in London, looking for lending opportunities. So we may have to change our geography a bit to get our volumes up. And if -- it would be -- I would tell you this, if I felt confident we could originate $15 billion in loans, up from $6 billion, we'd harvest all our gains in our equity books and put the assets out. Take the gains, take the -- and then because you -- that's equity that isn't returning a return -- that isn't earning a return, right?

It's a gain, it's not in our GAAP financials. When you do a screen on us, you see a 16-something book value. We know the drill, right? But nobody can say that service is a big problem today. I mean Rina mentioned to you that 90-something percent of our -- 95%, is it? It's 2.0. So the servicer has a $42 odd million book value now on -- in the company. It's worth probably double that or triple that. I don't even know.

Bigger.

So it's -- that's not -- we're complex. We talk about it every Board meeting, why we trade at the dividend yield we trade. For a while there, we seemed to be moving ahead. But we're going to stay the course. It's working for our shareholders reasonably well. It's not a huge issue for us. If the stock were $28, our dividend yield would be 5% or 6%, we could -- that would be really nice for us. We could issue equity accretively and go do great things. Right now, we don't have any need to do so.

And Europe looks like it's going to continue to be a good opportunity for us for a little while, and you're seeing some of our other competitors go there as well. The -- with rates higher in the U.S., the dollar is weaker on a forward basis. So given we hedge out the foreign exchange when we invest in Europe, you're picking up 200 to 230 basis points or so in the cross-currency swap today by investing in Europe, which makes dollar-denominated loans over there, brought back to dollars in the future look very attractive versus where they historically looked. So I would expect to see us continue to try to add in Europe. I think we brought the book back from -- it was 17-or-so percent of our book. It went down to 9%. I think it's back to 16-or-so percent today, and we would hope we can get it back to 20% or even above that over the coming quarters and years.

We have reached the end of the question-and-answer session, and I will now turn the call over to Mr. Sternlicht, Chairman and Chief Executive Officer, for closing remarks.

Well, we wish everyone a happy holiday season coming up. We'll speak to you again after the holidays. So have a great Thanksgiving with your families and a merry New Year and let's hope for stability and piece of the world. Thanks, everyone.

This concludes today's conference, and you may disconnect your lines at this time. Thank you for your participation.