Redwood Trust (RWT) Q3 2019 Earnings Call Transcript

Redwood Trust (RWT -1.56%)
Q3 2019 Earnings Call
Oct 30, 2019, 5:00 p.m. ET

Contents:

• Prepared Remarks
• Questions and Answers
• Call Participants

Prepared Remarks:

Operator

Good afternoon, and welcome to the Redwood Trust, Inc. third-quarter 2019 financial results conference call. [Operator instructions] Today's conference is being recorded. I will now turn the call over to Lisa Hartman, Redwood's senior vice president and head of investor relations, for opening remarks and introductions.

Please go ahead, ma'am.

Lisa Hartman -- Senior Vice President and Head of Investor Relations

Thank you, Gerry. Hello, everyone. Thank you for participating in Redwood's third-quarter 2019 financial results call. Joining me on the call today are Chris Abate, Redwood's chief executive officer; Dash Robinson, Redwood's president; and Collin Cochrane, Redwood's chief financial officer.

Before we begin, I want to remind you that certain statements made during management's presentation with respect to future financial or business performance may constitute forward-looking statements. Forward-looking statements are based on current expectations, forecasts and assumptions that involve risks and uncertainties that could cause actual results to differ materially. We encourage you to read the company's annual report on Form 10-K, which provides a description of some of the factors that could have a material impact on the company's performance and could cause actual results to differ from those that may be expressed in forward-looking statements. On this call, we will also refer to both GAAP and non-GAAP financial measures.

The non-GAAP financial measures provided should not be utilized in isolation or considered as a substitute for measures of financial performance prepared in accordance with GAAP. They are included to aid investors in further understanding the company's performance and to provide insight into one of the ways that management analyzes Redwood's performance. A reconciliation between GAAP and non-GAAP financial measures is provided in both our third-quarter earnings press release and Redwood review available on our website. Also note that the content of this conference call contains time-sensitive information that is accurate only as of today, October 30, 2019.

The company does not intend and undertakes no obligation to update this information to reflect subsequent events or circumstances. Finally, today's call is being recorded and will be available on the company's website later today. I will now turn the call over to Chris, Redwood's chief executive officer, for opening remarks and introductions.

Chris Abate -- Chief Executive Officer

Thanks, Lisa, and good afternoon everyone. Before I begin my remarks, I wanted to recognize the great efforts undertaken by our local officials, especially the firemen, evacuation coordinators, and others, who have bravely fought to contain Getty and various wildfires currently burning at this hour in California. These fires are very close to home for us, as a Northern California headquartered company, and we look forward to helping our communities recover and rebuild as these fires are contained. At Redwood, third quarter marked an historic time for us, were we made significant progress positioning the company for the future of housing finance.

Looking back to a few years ago, we announced a comprehensive new strategy to leverage our housing credit competencies across a broad and evolving residential landscape. This entailed not only the expansion of our consumer mortgage business, but also a commitment to offering similar solutions for housing investors to purchase residential real estate for business income. Since that time, we've demonstrated the skills and operations necessary to grow in this market, and have taken tangible steps toward building a best-in-class specialty finance platform, one that serves the liquidity needs of all homebuyers, homeowners and investors alike. Our expansion into the business purpose lending space began organically, but quickly evolved into a partnership with our 5 Arches team in Irvine.

The investment opportunities we have seen through 5 Arches have validated the thesis underpinning our new strategy. That is, the rate of new home building in the United States is not keeping up with the pace of new household formation. With our convictions strengthened, we further solidified business purpose lending as a core strategy at Redwood by recently announcing a second new partner in CoreVest. To recap our announcement from a few weeks ago, CoreVest is an industry-leading BPL originator with over 900 million of total assets and a profitable operating platform.

They bring an extremely talented team to Redwood that shares our values of working with integrity and fostering deep relationships with our customers and business partners. Since their inception in 2014, CoreVest has funded over 4 billion in loans, while developing industry-leading technology that offers a seamless borrower experience. The platform has generated attractive returns to date, with its highly scalable mortgage banking business significantly contributing to the company's profitability. The firm also recently completed its ninth securitization of single-family rental loans since it began issuing in 2015, more such transactions than any other issuer.

Pairing this channel with our own market-leading Sequoia program further extends our lead in the private label securitization market. It also strengthens our position as a leading lender in a market whose size we estimate to be in excess of 100 billion. Dash and Collin will provide more details about the near- and long-term benefits of this acquisition to our operating platform and investment portfolio in their prepared remarks. Moving on to this quarter's results, non-GAAP core earnings of $0.37 per share continue to comfortably cover our quarterly dividend of $0.30 per share, represented a $0.02 decline from the second quarter.

GAAP net income of $0.31 per share was up slightly from the second quarter, although our GAAP book value at September 30th declined about half a percentage point from the prior quarter. Obviously, we spent ample time in the third quarter focused on the CoreVest acquisition, our long-term funding needs, and our portfolio optimization goals, and some of these priorities contributed to a softening of our near-term results. In particular, our GAAP results included 1.7 million of one-time charges associated with our acquisition of CoreVest, an amount we excluded from our non-GAAP core earnings. From an operating perspective, the third quarter demonstrated our ability to leverage the strength of our business model to navigate through continued market volatility, particularly with respect to interest rates.

Low interest rates and fast prepayment speeds resulted in strong third-quarter mortgage banking volumes, although both hedging and securitization execution were challenging, as margins softened due to investors' uncertainty about prepaids going forward. Additionally, faster prepayment speed negatively impacted portfolio investments we hold at a premium, resulting in a temporary reduction in our non-GAAP core economic net interest income line item. Offsetting these effects was an improvement in the fair values for most of our subordinated investments as a result of structural deleveraging from prepayments. Additionally, mortgage banking margins for both securitization and home loan executions improved later in the third quarter and into the fourth quarter, as on-the-run, lower-coupon collateral made its way through the system.

Both the back-up in rates and the refi wave subsiding a bit have helped investors' appetites for AAA rated RMBS considerably. We ended up completing three profitable Sequoia securitizations during the third quarter, including our 100th overall, and continue to plan our normal course of deals in the fourth quarter and into 2020. Overall demand for residential mortgage credit has continued to strengthen throughout 2019 and drive asset prices higher for credit securities, and we continue to take advantage by rotating out of non-strategic investments, where our return expectations have been fully realized. With substantial portfolio optimization and the CoreVest acquisition now complete, we have increased our investment portfolio's sustainable earnings power going forward, and we expect economic net interest income to grow meaningfully, beginning with the fourth quarter of 2019.

Dash and Collin will have a lot more to say on our full operating results and optimization efforts, so I'd like to transition back to our core consumer residential business. Here, we remain committed to growing our expanded credit and non-qualified residential mortgage loan products by leveraging our approach to credit, speed to close, and reliable execution we deliver to our loan sellers. The third quarter marked a record quarter for Redwood choice purchases on strong quarter-over-quarter growth. We are very excited about the trajectory of the non-QM space, given housing finance reform initiatives by regulators over the course of this year, most recently the CFPB announcing in late July that they intend to let the so-called QM patch expire.

As we laid out in a presentation we published in May, the QM patch is a regulatory-driven competitive advantage afforded to the public mortgage sector that resulted in an unlevel playing field for non-QM mortgage lenders. We estimate the patch expiration will level the playing field for upwards of 185 billion of non-QM loans produced annually by Fannie Mae and Freddie Mac. To contextualize this opportunity, we purchased about 7 billion total of residential loans in 2018, a successful and profitable year for our consumer mortgage banking business. The success of our operating business has been directly complemented by the great work done within our investment portfolio.

The portfolio team continues to deploy capital at a record pace, leveraging unique and durable relationships, forged over several years. The key differentiator for Redwood has always been our ability to source and structure investments our competitors cannot easily duplicate. Our collective suite of businesses makes that truer than ever before, with over 672 million of capital deployed through October 30th, after including our acquisition of the CoreVest investment portfolio. As we grow our mortgage banking platforms, our portfolio activities and the efficiency of our corporate functions will be key to profitably scaling our consolidated business and maximizing the growth of earnings per share.

As we look ahead, credit-oriented strategies such as ours are in high demand, as the yield curve flattens and investors seek alternative ways to source real estate related assets. The product sourcing capabilities and operating know-how required to succeed, however, remain in scarce supply. Our company possesses a unique ability to bridge the gap between the customized needs of non-agency borrowers, whether BPL, non-QM, or traditional jumbo, and the liquidity options available to them in the marketplace. We're already making the necessary investments in technology and infrastructure to further automate our loan purchase process in anticipation of these opportunities.

With an eye toward integrating our consumer mortgage and BPL businesses over time across our national correspondent network, our vision of becoming the preeminent and most profitable specialty finance operator in the mortgage industry can be realized. And with that, I'll hand the call off to Dash.

Dash Robinson -- President

Thank you, Chris, and good afternoon everyone. Before I touch on results for the third quarter, I want to share more about our recent acquisition of CoreVest. As a refresher, our investment of $492 million completed earlier this month included CoreVest's origination platform, with an ascribed value of $130 million, and over $900 million of related financial assets, along with in-place financing. CoreVest's balance sheet is principally comprised of subordinate securities retained from the company's single-family rental securitizations, SFR loans held in pipeline slated for inclusion in upcoming securitizations, and short-term bridge loans made to investors seeking to stabilize or sell their portfolios.

As Chris mentioned, life-to-date, the platform has funded over $4 billion of loans, including $1.1 billion year to date through September 30th. Almost 70% of CoreVest's 2019 originations are comprised of single-family rental loans, a volume that is fed by both organic borrower cultivation and natural lead generation through the bridge lending book. As Chris noted, the acquisition of CoreVest reflects our strong belief that this area of housing credit offers substantial opportunity for growth and accretive returns for our shareholders, which we estimate to be in the range of $0.15 to $0.20 annually on a per-share basis over the course of the next 12 to 15 months. Our forecasted earnings accretion is a function of projected returns on the acquired financial assets, forecasted operating income of the origination platform, and flow of additional portfolio investments created through go-forward originations.

Importantly, this forecasted accretion does not fully encapsulate the acquisition's strategic rationale. An overlap in our competencies, most notably reliability and speed to close for clients and seasoned securitization platforms, is complemented by competitive advantages that we believe can bring unique to CoreVest's business. These include our extensive mortgage banking network, potentially fertile ground through which to expand product distribution, and a compelling cost of debt and equity capital. These are differentiators that we believe will drive additional value over time.

In addition to strategic fit, diversity and sustainability of revenue streams is important to emphasize in describing the value of the CoreVest acquisition and of our business purpose lending efforts overall. The expanded direct origination capability we now possess can create both long-term investments for our portfolio at attractive returns and a growing stream of fees and gain on sale from origination and securitization activities. We intend to hold the CoreVest bridge loans and subordinate SFR securities for investment in our portfolio, and expect to generate a 12 to 15% return on equity on these investments, taking into account associated in-place financing. Post-acquisition, over 15% of our investment portfolio is now allocated to BPL loans and securities, a concentration that we expect will continue to grow organically over time.

These expected returns on the financial assets do not include operating income generated by CoreVest's origination activities, which include loan origination fees and gain on sale, net of operating expenses. In the near term, we expect to generate a low to mid-teens core after-tax return on capital allocated to the operating platform, inclusive of both platform premium and working capital. Forecast returns on working capital in isolation, a pure reflection over time of the business's operating profitability, are substantially higher and a strong complement to our residential mortgage banking revenues. We believe having both the CoreVest and 5 Arches platforms under the Redwood umbrella positions us to scale profitably in what we view as a deep and growing marketplace.

As we discussed on our call announcing the CoreVest acquisition, the two platforms will not compete. In assessing CoreVest's business model, with 5 Arches already part of the Redwood family, we were intrigued by the distinct yet complementary ways in which the two platforms approached the marketplace. The key reason we find the BPL space compelling is what we believe to be a vast, untapped cohort of housing investors, whose financing needs are not being efficiently met. Many of these investors likely remain unaware that lending products are available to help them meet their financial objectives, whether it be in single-family rental, build-to-rent, or bridge and term lending, and single-family and small-balance multifamily.

While there is some overlap in approach, the platforms use unique means to access the market and develop fresh borrower relationships. This is evidenced by the relatively low incidence of borrower overlap we observed between the respective books of business. This is, in part, a testament to the depth of the market overall, especially given the maturity and size of each platform. We are excited to move forward and begin realizing on the cumulative origination of out business purpose lending platform.

Turning to our third-quarter results, the lower rates we saw in the first half of the year generally persisted throughout the third quarter, helping maintain residential loan purchase volume in our mortgage banking business, which totaled $1.5 billion. During the quarter, we completed one select securitization of $376 million and two choice securitizations totaling $727 million, and sold $470 million of whole loans to third parties. The mix of select and choice loan lock volumes shifted meaningfully, and third-quarter choice purchase volumes were up 37%. As Chris noted, the third quarter was a record one for purchases of choice loans.

The third quarter represented our second full quarter of ownership of 5 Arches. We have made meaningful progress on our integration efforts and have begun to see improved profitability across the operating platform and strong returns generated from assets held in our investment portfolio. Overall, BPL originations for the third quarter totaled $162 million, a small decline from the second quarter, driven largely by fundings for a number of loans pushing into early October. We anticipate improved run rates for the fourth quarter, and 5 Arches has already originated an additional $140 million of loans in October.

We remain pleased with the attractive risk-adjusted returns the 5 Arches platform has contributed to our business, and we see more opportunities to unlock value across the portfolio and the platform. Through the end of the third quarter, we have deployed $68 million of capital into over $200 million of bridge loans originated by 5 Arches, generating core returns on capital of approximately 11% for our investment portfolio. Optimizing this capital for our currently available financing terms, we estimate levered returns to increase to 12 to 15% over time. Additionally, we have begun financing 5 Arches' originated SFR loans using our FHLV financing facility, with an estimated return of 12 to 13%.

Our access to this low-cost financing is an example of one of our key competitive advantages and ability to increase long-term returns through optimizing how we use in-place financing capacity. Turning to our investment portfolio, capital deployment activity increased from the second quarter, as we continued to execute on unique third-party investment opportunities, combined with an increase on organically sourced investments. Specifically, our $152 million of deployment included $61 million toward our second investment in Freddie Mac issued subordinate securities backed by reperforming loans, or RPLs; an additional $21 million allocation to loans originated by 5 Arches, and $21 million into securities from our three Sequoia transactions. Investments in Freddie Mac issued RPL securities now comprise 11% of our total capital allocation and have represented a key strategic opportunity for our portfolio over the past several quarters.

Beginning in late 2016, as a complement to selling non-performing loans off of their balance sheet, Freddie Mac initiated securitization programs, transferring credit risk on seasoned single-family mortgage loans whose underlying terms had been modified from those at origination. In these structures, Freddie Mac has typically retained risk on the senior portion of the capital structure through either security ownership or credit guarantee, and has syndicated the subordinate securities to the private market. The structure provides subordinate noteholders an attractive senior financing rate locked in for 10 years, in contrast to alternative transactions in the marketplace, in which financing rates step up after three to five years. Central to the investment thesis in these securities is the view that borrowers will cure any persistent delinquency history and continue paying steadily under the modified terms of their loan.

This improvement in performance can sometimes be linked to the assumption of servicing duties by a high-touch servicer that specializes in working with borrowers to keep them current. Our estimated base case returns on these securities, which represent more than 20% of the transaction capital structure, range between 10 to 13% on a loss-adjusted basis, inclusive of the effects of leverage applied to the mezzanine tranches. The performance of our first Freddie RPL transaction continues to trend favorably, as delinquencies, typically representing borrowers that are one to two payments behind, have declined from 50 to 33% in less than one year. This decline in delinquencies has to date exceeded our originally modeled expectations.

Our capital-raising activity in the third quarter was elevated, given several opportunities we saw in the market. We organically generated investible capital through continued portfolio optimization activities, which included sales of securities and a strategic new financing transaction, which netted a combined $248 million of capital for redeployment, and in the process captured $25 million of realized gains. Security sales were focused in subordinate CRT securities, many of which we sold at substantial premiums to par value, and mezzanine multifamily securities. The recent debt facility structure has recourse to Redwood and is backed by a meaningful portion of our subordinate Sequoia securities.

The structure is non mark to market and has a three-year maturity with an option for two one-year extensions. The financing carries a coupon of 4.21% for the first three years. Now to recap our financial results, I'll turn it over to Collin.

Collin Cochrane -- Chief Financial Officer

Thanks, Dash, and good afternoon everyone. To summarize our financial results for the third quarter, our GAAP earnings were $0.31 per share, compared with $0.30 in the second quarter, and core earnings whatever $0.37 per share, compared with $0.39 in the second quarter. The increase in GAAP earnings in the third quarter was primarily driven by an increased benefits investment fair value changes from spread tightening in our security portfolio during the quarter. Core earnings per share decreased in the third quarter, primarily due to lower mortgage banking results and economic net interest income, which are both negatively impacted by rate volatility.

These decreases were partially offset by higher realized gains. While elevated portfolio optimization activity benefited gains during the quarter, growth in economic net interest income was dampened due to a higher average balance of undeployed capital, as we positioned ourselves for the acquisition of CoreVest and continue to be selective in redeploying capital due to the overall credit spread environment. Further, economic net interest income continued to be impacted by rate volatility during the quarter, which resulted in increased hedging costs and higher prepayments. While faster repayments helped to improve fair values of our subordinate investments, it negatively impacted portfolio investments held at a premium.

Our GAAP book value decreased during the third quarter to $15.92 per share which, along with our dividend, contributed to a 1.3% economic return for the quarter. While GAAP earnings exceeded our dividend during the quarter, book value decreased, primarily due to an $0.11 per share decline in the value of derivatives hedging our long-term debt, which were impacted by the decline in benchmark interest rates during the third quarter. Additionally, while GAAP earnings benefited overall by spread tightening on our subordinate credit securities, this benefit was partially offset by the negative impact of valuations from faster prepayments on our assets held at a premium, particularly our jumbo residential loans financed at the FHLB and our IO securities. Turning to the balance sheet and our capital position, we ended the quarter with approximately 590 million of available capital, exclusive of amounts set aside to repay our upcoming exchangeable debt maturity.

This was a significant increase from the second quarter, primarily due to an increase in capital-raising activity, which outpaced the capital we deployed during the quarter. During the quarter, we issued 220 million of common stock and freed up 118 million of capital from sales of low-yielding securities and 130 million from the new long-term secured financing facility Dash discussed. Subsequent to quarter-end, our acquisition of CoreVest utilized approximately 490 million of capital, and after other optimization deployment activity during the month, we estimate our current available capital to be approximately 100 million. Looking forward through the end of the year, we expect to continue portfolio optimization activities to free up capital for new investment opportunities, including new BPL assets created by our expanded platform.

And while we believe this will be sufficient to fund our near-term capital needs, as usual, we will evaluate the need for incremental capital in terms of what is in the best interests for our shareholders. In additional to our capital-raising activities, we also issued 201 million of six-year, 5.75% exchangeable debt, which is excluded from our available capital and will replace our current outstanding 201 million at 5.625% exchangeable notes, due in November. Looking at our capital structure holistically, our recourse debt to equity leverage ratio was 2.7 times at the end of the third quarter. The decrease from the second quarter was driven by our capital-raising activities and the timing of capital deployment.

In the fourth quarter, after deployment of capital, including for the CoreVest acquisition, we expect our leverage to increase to around three and a half times, within our long-term target range of three times to four times. Turning to our 2019 financial outlook, our results for the first nine months of the year, which included strong performance from our mortgage banking operations and elevated gains from portfolio optimization activities, have put us on track to generate returns near the high end of the overall range we initially provided in the beginning of the year. Breaking that down further, for the remainder of the year in our investment portfolio, we expect to see economic net interest income improve and realize gains to moderate. And in our residential mortgage banking business, we expect gross margins to normalize to within our long-term expected range and for volume to remain stable.

Now the section to review this quarter, we included some supplemental commentary around economic net interest income. To summarize, while we've experienced an overall decrease in ENII over the last two quarters, we expect this trend to reverse beginning in the fourth quarter and for ENII growth to continue into 2020. This will occur as we deploy our available capital into accretive investments, including those acquired from CoreVest, as recent investments begin to realize their full earnings potential, and as we continue to optimize our financing of several asset types. Additionally, we could see a meaningful benefit to ENII if prepayments normalize.

Finally, for business purpose mortgage banking platform, with the addition of CoreVest, we expect to increase our capital allocation to this business to around 10% of our total capital, and for the remainder of 2019, we expect to generate a core return on equity from the combined business purpose mortgage banking platforms of 12 to 14%. Further details on our outlook are provided in our third quarter 2019 Redwood review. And with that, I'll conclude our prepared remarks. Operator, please open the call for Q&A.

Questions & Answers:

Operator

Thank you. [Operator instructions] The first question is from the line of Douglas Harter, Credit Suisse. Please go ahead, sir.

Douglas Harter -- Credit Suisse -- Analyst

Thanks. Can you talk about the impact of higher prepayments on, you know, sort of your economic returns? So if you were to net, you know, kind of book value changes and earnings changes, kind of how higher prepays kind of flow through and impact you.

Chris Abate -- Chief Executive Officer

Sure, Doug. You know, there's a few different ways, and one thing we cited was we're pretty happy with the balance in our business. Obviously, on the mortgage banking side, volumes have been elevated, especially refi volume. That said, we saw some really fast prints starting in June and July, particularly the 2018 vintage loans, and that definitely impacted securitization execution in the third quarter.

I think investors, as we moved into some of the lower-coupon products, have definitely regained a lot of comfort, and that execution has strengthened a lot later in the third quarter, into the fourth quarter. You also had the impact on our premium investments, which we noted. I think all told perhaps in the neighborhood of $0.03, give or take, was probably the net impact. But, again, if you're in the mortgage banking business, you've got to keep the loans moving, so we're pleased to have gotten off the three deals that we did.

And as I mentioned, we have no plans to alter the pace of activity going forward.

Douglas Harter -- Credit Suisse -- Analyst

Great. And then, I think you mentioned kind of the level of execution profitability kind of improved going, you know, as the quarter was ending. Can you just talk about how October looked, from kind of a mortgage banking profitability standpoint?

Dash Robinson -- President

So, Doug, it's Dash. We did complete one deal in October, and if you compare the execution of that transaction to what we did in the third quarter, the execution on the AAAs improved by about three-eighths of a point in price, which was a function of a couple things, of course, where rates have gone, but also, to Chris's point, the ability to securitize collateral whose coupon was closer to current production coupon certainly helped. The execution on that transaction was quite strong on the AAA securities, and actually, the tightest we'd seen in several years, so the market has certainly bounced back here in October. It's certainly somewhat driven by demand, but also the normalization of rates and what we're securitizing versus where mortgages are trading generally.

Douglas Harter -- Credit Suisse -- Analyst

Right. Thank you.

Operator

The next question is from Stephen Laws, Raymond James. Please go ahead, sir.

Stephen Laws -- Raymond James -- Analyst

Thank you. Good afternoon. Collin, I guess kind of following up on your comments, and I appreciate the update, it sounded like pro forma capital at the end of October was roughly around 100 million. Can you talk about, maybe little more detail on any additional lower-yielding securities you guys have identified that you'll look to sell down as other investment opportunities present themselves, as well as optimization efforts you can make on financing business purpose loans in some capacity that might be more attractive than the current warehouse facilities you have in place?

Collin Cochrane -- Chief Financial Officer

Yes, you know, in terms of optimization, I think we've made a significant amount of progress this year, but I think there still remains a little bit of work to do there. We still have some CRT investments that I think we're looking at as potential candidates for optimization, and potentially some multifamily mezz that we're looking at. So, I still think, heading into the fourth quarter, I mentioned that there is a little bit more optimization that we expect to do, but I do think we have made a significant amount of progress year to date. So, we do see that optimization sort of tapering down through the end of the year.

And in terms of financing, we did, again, make some good progress this year putting -- or this quarter, rather, putting this new term loan in place. In terms of some of the financing on our business purpose loans, I think there's a little bit of work we expect to do there through the end of the year to tighten up some of the terms and expect to see some improvement. And then, Dash also mentioned some work we've been doing to get SFR loans eligible to refinance over on the FHLB. So, we moved a small balance of loans over before the end of the third quarter, and we're looking to get the substantial majority of the remaining SFR loans that we originated through 5 Arches over in finance this month for the end of the year.

So, I think those are really some of the more immediate things that we're looking at over the course of the next couple months here, heading into year-end.

Dash Robinson -- President

Stephen, it's Dash. One thing I would add to that, and we talked a bit about this on our call a couple weeks ago announcing the acquisition, you know, as you know, we inherited a fair amount of financing in acquiring CoreVest, and there are, from our perspective, some near-term things we can do to sort of normalize that structure into how we typically structure those sorts of warehouses, which we hope in the relative near to medium term can be more of an immediate impact to the returns on financing. Because the assets, I think Collin covered the waterfront really well, probably, but I just wanted to reinforce that point that we mentioned a couple of weeks ago, because that could potentially be very meaningful.

Stephen Laws -- Raymond James -- Analyst

I appreciate the color from both of you, and thanks for the reminder on that weeks ago. And, Dash, I was writing down as quickly as I could, but I want to circle back. Did you say $0.15 to $0.20 accretive on CoreVest, or was that a 15 to 25% ROE versus low to mid-teen ROE? I didn't quite catch the 15 to 20 that you mentioned. I was writing something else down.

Dash Robinson -- President

Sure, Stephen, no problem. Yes, there were a fair amount of numbers and percent hedges in the script. The accretion was $0.15 to $0.20 per share per annum, which is based upon assumptions about what we've purchased, of course, and then assumptions around the go-forward profitability, and then creation power of the platform versus alternative uses of capital. The return profiles that we've stated are in the 12 to 15% range, which are certainly a big part of driving that accretion calculation that we talked about.

Stephen Laws -- Raymond James -- Analyst

Great, thanks for clarifying that. Last question for me here. Chris, could you maybe talk higher level GSE reforms? Specifically, Collin mentioned, and Dash as well, FHLB financing, but I know one proposal might open that back up again, you know, allow you guys to have maybe more access there. And then, secondly, comments around possibly a new qualifying mortgage definition, and maybe what the latest is you're hearing around those issues, and anything else that you'd like to highlight on potential GSE reform.

Chris Abate -- Chief Executive Officer

Thank you, Stephen, I hope you're doing well. There's obviously a lot going on. The SEC actually just released something today requesting feedback from the market on Reg AB II, risk retention, and some of the issues, the gating issues that have prohibited or been an impasse for the private label sector to do more public transactions, so we've been pretty active on that front. We published some pretty meaningful content, I think in August, on ways to do that, lessons learned from the 144A market and so forth, so that's one thing that is very current.

You know, I think on the QM patch front, we've had some recent statements from the director, sort of reaffirming the intent to allow that to expire. I think the private sector needs to step up and do its part. You know, we certainly plan to do ours. We're working quite a bit internally with automation efforts and technology, to try as best as we can to make the transition as painless as possible for originators.

That said, we feel very strongly about a level playing field. We feel like it's certainly in the interest of taxpayers. We do feel like being able to compete on the same front will definitely move a significant amount of these loans away from taxpayers into the private sector, so we're excited about that. We've also been somewhat vocal on risk retention and some of the issues there.

From a practical perspective, one loan, one non-QM loan in a QM securitization triggers full risk retention, so we've advocated things such as asset-specific risk retention and some other alternatives that will start to blunt this binary trigger effect when a loan goes from QM to non-QM. We did provide some thoughts on the QM definition, and our perspective, we still believe that credit metrics matter. We don't want to move fully to market-based metrics or definitions for QM versus non-QM. I think what we've emphasized is, wherever that definition lies, whether it's 43 DTI or 50 DTI, a big challenge in the market is the drop-off, and wherever it occurs, whether it's 43 to 44, 50 to 51, or some other metric, we're looking for ways to make that drop-off less acute.

So, there's a lot going on. There's going to be some more announcements here in the coming months, but we definitely are pressing forward and are certainly looking to step up and absorb as much of this volume as we can.

Stephen Laws -- Raymond James -- Analyst

Great. Well, between the two acquisitions this year and the upcoming reform proposals, certainly seems like next year will be quite an exciting year. Appreciate you taking my questions today.

Operator

The next question is from Steve Delaney, JMP Securities. Please go ahead.

Steve Delaney -- JMP Securities -- Analyst

Hey, folks. Look, first, before questions, just sending prayers to you, your families, and all your communities to get safely through these fires. It's just a terrible thing, and it seems to happen repeatedly, so we're thinking about you. On CoreVest, I mean, obviously it's going to take us -- after a quarter, so you guys will -- you won't have to listen to us ask CoreVest questions, but for now, we need to.

Did you ever mention publicly what your headcount was at CoreVest? I recall 5 Arches was like 90 people or something, but do you have a figure you can share on CoreVest so we can get a sense for how large the platform is.

Chris Abate -- Chief Executive Officer

Sure, it's about 100 FTEs.

Steve Delaney -- JMP Securities -- Analyst

About 100, and was I correct on 5 Arches at 90, or was that too high? I don't know why I had that in my mind.

Chris Abate -- Chief Executive Officer

That was about the number when we completed the acquisition of the platform in March. It's a little over 100 now, just with additions across the platform --

Steve Delaney -- JMP Securities -- Analyst

OK.

Chris Abate -- Chief Executive Officer

Facilitate more growth. So the FTE count of each platform is actually pretty close.

Steve Delaney -- JMP Securities -- Analyst

OK, great, that's helpful. Now, third quarter, you guys had consolidated operating expenses at $25 million. Collin, I think it was down a million from 26. Now, CoreVest obviously didn't close until fourth quarter.

I'm going to come back to you -- you know, thank you for the $0.15 to $0.20. But just as far as the overhead of the operation, ignoring all revenues, do you have a sense for on either on a quarterly or an annual basis what CoreVest specifically would add to the to the and load on a quarterly basis?

Collin Cochrane -- Chief Financial Officer

Yes, I think on a quarterly basis, we're looking at about 5 to 6 million of expenses as probably reasonable run rates.

Steve Delaney -- JMP Securities -- Analyst

OK, great. That's very helpful. Thank you, Collin. And, Dash, in your remarks, what I took it to mean, you know, you talked about the loans coming out of CoreVest on your balance sheet.

You're looking for a 12 to 15% return on equity. But since you mentioned, you know, without fees and gain on sales. And I understand you guys are trying to run this as like a, it's part of Redwood, but it's a stand-alone business that needs to stand on its own two feet. And it sounds -- you're leaving fees, origination fees, and any gain on sale in CoreVest to help offset expenses.

Is that what I heard you say?

Dash Robinson -- President

Yes, I appreciate that question, Steve, because it brings up an important point about just as we -- look at the value creation of these platforms holistically, you know, the actual operating metrics, revenue less expense, and the value of the loans and securities created. You know, some of this is a function of the just the geography of where the businesses sit. CoreVest, like 5 Arch, that operating platform is at our TRS because of the functions that it does day to day and the origination --

Steve Delaney -- JMP Securities -- Analyst

Sure.

Dash Robinson -- President

And so, you know, those fees and the gain on sale associated with securitizing SFR loans and the borrower points associated with originations, similar to 5 Arches, those geographically will be at the TRS, and of course the long-term investments will go to the REIT on a more tax-advantaged basis. But, you know, similar to my remarks on 5 Arches, with CoreVest, we look at and measure performance holistically as it relates to the assets created and the operating metrics, you know, those two go hand-in-hand. And so, some of it is just split geographically, you know, based on our tax structure, frankly. But when we assess performance and measure ourselves, it's holistic with assets and then the operating metrics together.

Steve Delaney -- JMP Securities -- Analyst

OK, that's helpful. And when you -- you mentioned gain on sales tied to the securitizations, you know I think at one point, I can't remember if it was 5 Arches or CoreVest, somebody was selling off their bridge. But is -- the current situation, are you essentially retaining all production and then not selling anything off on a whole-on basis, but you retain it and then you choose to securitize it or put it with FHLB or whatever. Is that -- I mean, I guess what I'm saying is, is everything that they're producing coming to Redwood, or are you actually trying to sell some of it away?

Dash Robinson -- President

That is predominantly the case. With the 5 Arches, we're to the point where we're retaining essentially 100% of what that platform produces. There are a couple of exceptions, but there's no front-foot effort there to sell loans to third parties. With CoreVest, there's a small part of the production which is sort of a smaller balance SFR product, which to date, the platform has been selling.

That's partly what I was alluding to as a product that could potentially be very logical to marry up with our mortgage banking network, and so the options are sort of open to us at this point. We haven't concluded exactly yet the right outcome there. We could certainly keep them and securitize them, along with the more regular-way SFR loans. But for the most part, the intention is to retain those only like we talked about before, similar to mortgage banking.

It's important to keep the muscle memory open to be able to sell these loans to counterparties, depending on risk return, where the market's pricing, risk, etc.

Steve Delaney -- JMP Securities -- Analyst

OK. Well thank you all for your comments, and I look forward to seeing you in a couple weeks.

Dash Robinson -- President

You, too. Thanks.

Operator

We have a question from Bose George, KBW. Please go ahead, sir.

Bose George -- KBW -- Analyst

Hey, guys. Good afternoon. Going back to your 2019 guidance, did you narrow that down to a core ROE range, or was it really just on the line items? I can't remember.

Collin Cochrane -- Chief Financial Officer

Yes, I mean, we've typically expressed our guidance in terms of components where we allocate our capital between the businesses, and so that's the format in which we've laid it out, and that's what we updated things relative to in the fourth quarter. So we broke that down between our investment portfolio, our residential mortgage banking business, and then we gave an update on the business purpose component of mortgage banking, inclusive of CoreVest. And so, that's what we've laid out in the regulatory review.

Bose George -- KBW -- Analyst

OK. And then, I was trying to think, the $0.15 to $0.20 accretion, you know, should we think of that really on top of the core earnings number that you guys end up with in 2019, and it's $0.15 to $0.20 over that? Is that kind of the right way to think about it?

Chris Abate -- Chief Executive Officer

Yes, going forward. You know, we've offered EPS guidance as a firm in the past, and we have in the past few years, but we felt like since we just completed this acquisition, we felt like we needed to do something more holistic. And so, I think we'll think about how to incorporate that guidance into next quarter's review, where we refresh our annual guidance, but for now, that's sort of in addition to what we've currently been earning.

Bose George -- KBW -- Analyst

OK, thanks. And then, in terms of the dividend, you guys had talked in the past about increasing the dividend closer to core earnings. I'm just curious, now you have obviously more earnings at the TRS through the acquisition. You know, just updated thoughts on how you view that.

Chris Abate -- Chief Executive Officer

Yes, I mean, what we're trying to do, and what we -- we spent a little bit more time this quarter talking about economic net interest income, because we've put a lot of the puzzle pieces together, we think, to drive that -- those durable, sustainable cash flows. We've been optimizing, and those are one-time realized gains, per se. But I think what we've done is we've really extended our runway and the line of sight and how we get to core and the consistency of core. We feel like we've put a portfolio together that will be a little bit more transparent in how we generate cash flows.

Ultimately, of course, the goal is to have those consistent durable cash flows that will enable us to raise the dividends, so that's something we're focused on.

Bose George -- KBW -- Analyst

OK great. Thanks.

Operator

We have a question from Matthew Howlett, Instinet Nomura. Please go ahead, sir.

Matthew Howlett -- Nomura Instinet -- Analyst

Hey, Everyone thanks for taking my question. I apologize if I missed it, but did you give an origination guidance number for CoreVest, or just single-family rental combined, business purpose loan combined for 2020?

Dash Robinson -- President

Matt, it's Dash. No, we didn't. We'll have more to say on that when we get a quarter behind us here. You know, I would sort of point you to the current run rate of production, which we've talked about.

I can certainly see room for growth, but we haven't, at this point, put a number.

Matthew Howlett -- Nomura Instinet -- Analyst

OK. Just help me think about the securitization process. How big are the retained interests that you would take down in terms of the subordinate tranches? And then, remind me a little bit of what the collateral -- I mean, these are prepayment penalty loans, are they not, and the coupons are, what, maybe six-plus percent?

Dash Robinson -- President

Yes, so taking those in reverse, yes, these loans typically have five- to 10-year terms, either or 30-year and/or some IO period. And yes, it's certainly one of the compelling things about them is, from a commercial real estate like perspective, they tend to have strong prepayment protections. The vast majority of what CoreVest securitizes has yield maintenance typically up through six months prior to the maturity of the loan, or there's a declining point structure, but yield maintenance is much more common. You know, there are certainly some loans with rates at 6%, but with rates having declined as much as they have, the average coupon is in the fives.

At this point, that will obviously move around with rates. From a securitization perspective, you know, in terms of retention, we estimate retaining, all in, 6 to 8% plus of the securitization. That would include the subordinate credit securities, as well as potentially some of the interest-only securities that are part of the structure. So, at a high level, not unlike Sequoia, although the thickness of what we're retaining, which is part of the attractiveness, obviously at those return targets is significantly higher than what we're typically retaining on a select transaction -- and even, to an extent, choice.

Matthew Howlett -- Nomura Instinet -- Analyst

Yes, that was my next question. Redwood practically amended the jumbo market. These structures that are coming out, is there anything relatively more attractive about investing in these securitizations versus what you've done traditionally in the jumbo space?

Chris Abate -- Chief Executive Officer

Yes, I mean, I think Dash hit the nail on the head in the sense that we can just put more capital to work at similar or better returns. I like to liken this market to where the jumbo space was 10 or 15 years ago. And so, when we think about our role in the markets as a specialty finance company, we're really meant to be involved here and take some of the lessons we've learned and experience we've gained from jumbo and apply them here.

Matthew Howlett -- Nomura Instinet -- Analyst

Got it. My next question, I mean, you guys, you've grown the choice program well, you're securitizing, and you have that bucket where you show sort of the third-party issuance that you take, and that's been sort of flat for a while. As we start to see these non-QM conduits come out, these banks that even do prime jumbo, I mean, how willing will Redwood be investing in third-party non-QM deals or anything that comes out from the banks, given expected GSE reform? I know in the past, Redwood's been pretty aggressive with some of the big banks. I was just curious how you're viewing third-party deals today, with some of the originations you're creating.

Chris Abate -- Chief Executive Officer

We've shied away, but really because we haven't liked the value proposition. So, as we've said, we've been doing a lot of optimization because asset prices have gotten so high. It's been great for gains, but not so great for putting capital to work. It's one of the reasons why we've progressively gotten more organic and are focused on sourcing the raw material to create the deals, rather than the deals themselves.

That said, you know, if there's a significant increase in supply, that should have a countering effect on pricing. So, we're always open and looking at third-party opportunities we see in the market, and we can certainly see, with that supply going up, the opportunities becoming more attractive.

Matthew Howlett -- Nomura Instinet -- Analyst

Great, thanks. And the last question's on, you know, you've been watching, Chris, a lot. There was a comment on maybe letting the REITs active regain access to captive reinsurance. I know you're FHLB line doesn't expire until 2026, so just curious what you're hearing on that front.

Chris Abate -- Chief Executive Officer

Sure, and I probably should have addressed this earlier with Steve, and I apologize. But there's a lot going on, some of which is not ready for prime time, so to speak. But I can say that the new regime at the FHFA is very aware of it. I've made them personally aware that the QM patch expiration set for January of 2021 is right on top of the stated expiration of the grandfathered captives in January of 2021.

So, we've been very clear that if they're moving that line of liquidity for the folks in the private sector, that they need to step up to assume some of this business from the GSEs is not going to be helpful. I think the real essence of the solution resides in safety and soundness for the system, first and foremost, and ensuring that the infrastructure is taking into account, whether it's captives or other forms of members. I think if there can be something that is compelling and ensures the safety and soundness of the system, hopefully, the regulator is receptive. But at this point, candidly, I think that the FHLB captive issue is a little bit behind the QM patch and some of the, you know, the GSE privatization initiatives, or removing them from conservatorship.

But those issues are, I think first and foremost, but this is quickly gaining some, I think, more greater notoriety within Washington, and we're certainly advocating for a solution.

Matthew Howlett -- Nomura Instinet -- Analyst

Great. Thanks for that.

Operator

We have a follow-on question from Bose George, KBW. Please go ahead, sir.

Bose George -- KBW -- Analyst

Great, thanks. I just wanted to follow up on the question about the -- Matt's question just on the yield of the portfolio. Actually, how does it vary between the BPL and the SFR loans? And also, I thought the coupon at least on the existing bridge portfolio I thought was quite a bit higher, so I just wanted some color on that.

Chris Abate -- Chief Executive Officer

Sure. Bose, just to clarify, your first question is returns on the BPL versus Sequoia or --

Bose George -- KBW -- Analyst

No, just the SFR, the single-family, the investor stuff you're doing there as well.

Chris Abate -- Chief Executive Officer

Yes, so in my response to Matt in terms of the coupons was specific to SFR loans. The average coupon for Bridge is about 8%, and that doesn't include the borrower points at the platform or origination, but those loans are 200 to 300 basis points higher in coupon than SFR.

Bose George -- KBW -- Analyst

OK, great. Yeah, I just wanted to confirm that. Thanks.

Operator

[Operator signoff]

Duration: 61 minutes

Call participants:

Lisa Hartman -- Senior Vice President and Head of Investor Relations

Chris Abate -- Chief Executive Officer

Dash Robinson -- President

Collin Cochrane -- Chief Financial Officer

Douglas Harter -- Credit Suisse -- Analyst

Stephen Laws -- Raymond James -- Analyst

Steve Delaney -- JMP Securities -- Analyst

Bose George -- KBW -- Analyst

Matthew Howlett -- Nomura Instinet -- Analyst

More RWT analysis

All earnings call transcripts