MFA Financial, Inc. (MFA) Q4 2019 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by, and welcome to the MFA Financial Fourth Quarter Earnings Conference Call. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session. Instructions will be given at that time. [Operator Instructions] As a reminder, this conference is being recorded. I'd now like to turn the conference over to our host, Mr. Hal Schwartz. Please go ahead sir.

Thank you, operator. Good morning, everyone. The information discussed on this conference call today may contain or refer to forward-looking statements regarding MFA Financial Inc., which reflect management's beliefs, expectations and assumptions as to MFA's future performance and operations. When used, statements that are not historical in nature, including those containing words, such as will, believe, expect, anticipate, estimate, should, could, would or similar expressions are intended to identify forward-looking statements. All forward-looking statements speak only as of the date on which they are made. These types of statements are subject to various known and unknown risks, uncertainties, assumptions and other factors, including those described in MFA's Annual Report on Form 10-K for the year ended December 31, 2018, and other reports that it may file from time-to-time with the Securities and Exchange Commission. These risks, uncertainties and other factors could cause MFA's actual results to differ materially from those projected, expressed or implied in any forward-looking statements that it makes. For additional information regarding MFA's use of forward-looking statements, please see the relevant disclosure in the press release announcing MFA's third quarter 2019 financial results. Thank you for your time. I would now like to turn this call over to MFA's CEO and President, Craig Knutson.

Thank you, Hal. Good morning, everyone. I'd like to thank you for your interest in and welcome you to MFA Financial's fourth quarter 2019 financial results webcast. With me today are Steve Yarad, our CFO; Gudmundur Kristjansson and Bryan Wulfsohn, our Co-Chief Investment Officers; and other members of senior management. 2019 was a very good year for MFA and for our shareholders. Total shareholder return for the year was 27.5%. Our book value was very steady for the year with the largest quarterly change in book value of less than 0.75%. We introduced the concept of Economic book value, which captures the fair value of our loan portfolio accounted for at carrying value for GAAP. These loans represent nearly half of our investment portfolio, and the fair value of these assets is approximately $180 million higher than their carrying value. Also during 2019, we refreshed our at the market or ATM program, which had been dormant for almost 10 years. This offers us the ability to sell shares into the market from time-to-time as market conditions permit. We issued our first convertible bond offering in May, raising $230 million and adding another capital raising tool to our repertoire. We're also proud to report that MFA was one of 325 companies named to the Bloomberg Gender-Equality Index. This is a select group from nearly 6,000 companies across 84 countries and regions and 50 industries. As we close the books on 2019, it's hard to imagine how different the fixed income market environment was a year ago versus today. At the beginning of 2019 the 10-year treasury yield was 2.70. The Fed had raised rates in late December and the market was generally expecting a continuing tightening cycle. Instead, the Fed paused in a very public way with the release a year ago of the minutes of their January meeting, and rates drifted lower through the first half of the year with the 10-year just over 2% by the end of June. The Fed then cut rates on August 1, and rates rallied sharply with the 10-year yield dropping to as low as 1.45 by the end of August. Markets experienced a bout of surprising illiquidity in the overnight funding market for treasuries and Agency MBS in mid-September, but decisive Fed intervention and two more rate cuts in September and at the end of October settled the markets and rates have been relatively stable for the last few months. Markets now seem to anticipate a relatively benign interest rate environment with the Fed on hold for the foreseeable future. For levered investors in mortgage assets, this environment presented some formidable tests, but we're delighted to report that MFA continued to thrive despite these challenges. MFA's investment strategy is very much intentionally not dependent on accurately predicting interest rate moves and these unexpected market developments had no material negative impact on our financial results. MFA's investment acquisition strategy particularly our focus on purchased performing loans in which we include non-QM, fix and flip, and single-family rental loans is proving to be a durable model. The groundwork that we laid beginning in early 2017 gained further traction in 2019 as our origination partners grow their businesses at least in part through MFA's continued appetite to purchase loans. MFA's reputation as a reliable buyer of residential whole loans and dependable capital partner has enabled us to source significant volume of whole loans. Despite the difficult mortgage environment, we continue to make investments that provide solid returns on equity. The economic backdrop is constructive for our residential mortgage credit focus as the U.S. economy remains resilient, labor market is strong, and housing stable. Please turn to page 3. MFA's GAAP earnings per share was $0.21 in the fourth quarter and we paid a $0.20 dividend to common stockholders on January 31. MFA has paid a $0.20 dividend now for 25 consecutive quarters. Core earnings was also $0.21 in the fourth quarter. We acquired $1.7 billion of assets in the fourth quarter of 2019, including $1.5 billion of whole loans. Our whole loan investment portfolio increased during the year by $2.9 billion with our overall portfolio increasing by over $1 billion for the year. Our book value decreased slightly to $7.04 at the end of the year from $7.09 at September 30, but our Economic book value increased slightly from $7.41 to $7.44 per share. Our economic return for the quarter was 2.1% under GAAP and 3.1% using Economic book value. Please turn to page 4. The fourth quarter of 2019 was our most active investment quarter of the year as we purchased approximately $1.7 billion of assets including $1.5 billion of purchased performing whole loans. New investments in 2019 totaled $5.3 billion including $4.3 billion of whole loans. The process of acquiring newly originated performing whole loans is very different from that associated with our other asset classes as we generally purchase these loans directly from originators rather than from The Street or through bulk offerings. Through our willingness and ability to explore various arrangements including flow agreements, strategic alliances and in some cases capital investments in originators, we've been able to partner with these entities to source attractive new investments while enabling them to grow with support from MFA as a reliable capital provider. These investments in originators range from equity purchases in all cases minority stakes to preferred stock investments some of which include warrants for common equity to convertible debt instruments. This strategy is consistent with our approach whereby we provide capital in the form most suitable for our origination partners seeking additional capital. During the fourth quarter, we increased our aggregate investments across five origination partners to $148 million. And during the quarter these investments generated interest and dividend income of $3.7 million before allocation of profits. Please turn to page 5. As we have discussed previously, our expanding investments in newly originated loans or purchased performing loans is beginning to have a meaningful impact on our interest income. These loans are included in our loans held at carrying value on our balance sheet. Recall that we also include loans purchased as reperforming loans or purchase credit impaired loans in our loans held at carrying value. In the fourth quarter, all loans at carrying value produced $72.3 million of interest income. This is compared to $101 million for all of 2018 and up from $64.2 million in Q3 of 2019. As we continue to grow our balance sheet, we will add marginally more leverage particularly on our residential whole loans held at carrying value. You will note that our overall leverage ratio ticked up in the fourth quarter from 2.8 times to 3.0 times as we added approximately $1 billion of additional borrowing on whole loans. And despite the fact that this whole loan borrowing is generally our most expensive financing we nonetheless saw total interest expense decline in Q4 versus Q3. This was due to lower LIBOR levels following Fed rate cuts during the third quarter. One-month LIBOR was approximately 2.25 before the Fed rate cut on August 1, but it was still around 1.75 even after the third rate cut on October 31. And LIBOR actually ticked up to about 1.80 in late December before finally settling out in the mid 1.60s recently. Since our whole loan repo is contractually tied to LIBOR and generally resets monthly we should continue to experience some tailwinds as these resets work through the lower rates. We also expect to begin to tap securitization markets particularly for our non-QM loans, which should provide an additional funding benefit and offer moderately higher levels of leverage. Please turn to page 6. Through our credit-sensitive whole loans, we've committed significant resources to our asset management efforts. We recognized that by immersing ourselves in the complicated and sometimes messy details of managing credit-sensitive loans that we can achieve better outcomes and improved returns. As good as our third-party services are there is a tangible benefit to direct oversight and involvement in decision-making. Our asset management team not only oversees our loans purchased as non-performing loans including the REO properties that we've acquired as a result, but also other portfolio loans that become delinquent. We believe that this expertise will serve us well and will drive better and faster resolutions. Our information technology team has also played a very active role in improving processes and efficiencies as our loan portfolio continues to grow. We had a number of our Legacy Non-Agency bonds that were called in the fourth quarter boosting our yield on this asset class for the quarter to 14.8%. Approximately, 70% of this portfolio is currently callable or expected to become callable during 2020. And finally, we purchased an additional $93 million of MSR-related assets in Q4 of 2019. This portfolio has doubled in the last year and now stands at $1.2 billion. Please turn to page 7. To summarize our strategy and initiatives for 2020, we expect to continue to increase our investments in purchased performing loans, specifically non-QM, fix and flip and single-family rental. When and if we're able to grow our other existing asset classes at attractive levels, we will obviously continue to do so. And as always, we are constantly evaluating new investment opportunities. Given our track record we are usually among the first to see new opportunities as we have demonstrated in the past the ability and willingness to help structure these deals and invest in size. As previously mentioned we'll look to optimize our capital structure around our assets through securitizations which will diversify our funding and provide some modest cost savings. We'll likely continue to execute strategic sales of Legacy Non-Agency MBS. This is part of managing a mature portfolio and includes sales of bonds at relatively high prices with little additional upside, sales of callable bonds at a premium and sales of low loan count or odd-lot position sizes at attractive round lot levels. We continue to manage our CRT portfolio by selling seasoned deals trading at very tight spreads and high dollar prices and adding newly issued CRTs when pricing is attractive. Please turn to page 8. We are excited about the opportunities ahead in 2020. We think we are well-positioned to continue to source newly originated loans through our origination partners most of whom are growing. A benign rate environment can assist in this growth as loan brokers tend to migrate to these more time-consuming originations as the easy refi loans become less viable. Capital structure optimization together with lower rates should provide a marginal earnings tailwind. GSE reform is a constant topic of discussion and debate. And while we have no unique insight as to what GSE reform looks like it's very difficult to envision a scenario in which GSE reform does not open up new opportunities for MFA to make attractive investments. And now I'll turn the call over to Steve Yarad, who will provide further details on the financial results for the most recent quarter.

Thanks, Craig. For the fourth quarter of 2019, MFA's net income to common shareholders was $96.9 million or $0.21 per share. GAAP earnings comfortably exceeded the prior quarter, and again covered dividend distribution. Growth in our investment portfolio, GAAP net income higher in each quarter, of 2019, in addition core earnings, which excludes the impact of unrealized gains and losses, on certain investments in residential mortgage securities and related hedges, that are included in GAAP earnings each period, was also $0.21 per common share. Since adopting the core earnings metric in, Q1 of 2019, MFA has delivered very consistent and steadily improving earnings performance, with minimal differences between our GAAP core quarterly results. This is largely the result of the successful execution of our investment strategy and expansion of our investments, in residential whole loans. Please turn to page 9, where we present additional information and the highlights, of MFA's net income for this quarter, which were as follows. Net interest income was $0.03 per common share higher than the prior quarter and -- one, the impact of the early bond redemption, particularly Legacy Non-Agency MBS, which resulted in the recognition of an additional $11.6 million of interest income, this quarter. While the level of fund redemption and the impact on this quarter's results was higher than we have experienced in prior periods, it should be noted that this quarter's net interest income, would have been sequentially higher, even if these bond redemptions had not occurred. Two, continued growth of our residential whole loan portfolio, resulted in a 15.7% sequential quarter increase, in net interest income. In addition, net interest income from residential whole loans, increased in each quarter of 2019 and 2018. And three, as Craig noted, and as Gudmundur will discuss in more detail in the upcoming slides, reductions in LIBOR are having a positive impact on funding costs across our portfolio, with net cost of funds falling by approximately 17 basis points from the prior quarter. Other income was approximately $0.02 per common share, lower this quarter, as fewer sales of residential mortgage securities, resulted in smaller realized gains. However, our loans accounted for at fair value, continue to make a significant contribution to our results. As was the case in the previous quarter, 60% of net gains on residential whole loans measured at fair value reflect coupon income receipts, gains on loan liquidations, and other [Technical Difficulty] Finally, operating and other expenses were $1.5 million higher, this quarter. Despite modest overall delinquency levels, the growth in our performing loan portfolio has resulted in a need to record additional provisions, for credit loss. We also incurred higher expenses related to management of our REO portfolio, as it continues to grow. Our G&A expense was essentially flat this quarter, at 1.5% of stockholders' equity, and continues to be in line with our expected run rate. Turning to slide 10, we present MFA's full year results for 2019 and 2018. What is noteworthy here is that nearly 12% year-over-year increase in net interest income, which was fueled by growth of our residential whole loans, as I previously discussed. In addition, other income was substantially higher, anchored by the consistently strong performance of our fair value loans. Further, the relatively less volatile market conditions we experienced in 2019, in combination with lower holdings of Agency MBS and CRT securities, accounted for at fair value, resulted in a reversal, prior year unrealized losses on these investments. Please turn to slide 11. We thought it would be helpful to provide an overview of the key impacts on MFA, of the new current expected credit losses with CECL accounting standard. We've been working hard for many months to be ready to implement CECL. And I'm happy to report that MFA adopted the new standard at the beginning of this year. The standard has no impact on our 2019 financial statements. But we were required to record the transition impact of adoption, as an adjustment to stockholders' equity, on January 1 2020. As a high level overview, CECL adoption primarily affects how we estimate credit losses, on our residential whole loan investments. Under CECL, we are required to make life-of-loan estimates of expected credit losses, using a combination of future forecasts and historical loss experience. This approach is very different from the so-called incurred loss model, used prior to 2020, where reserves were booked only if credit losses were assessed as probable, of being incurred. Consequently, under CECL it is reasonable to expect, that estimates of credit losses will be higher and recognized earlier, and would have been the case under the new credit loss approach. On slide 11, we summarize the key impacts of CECL adoption on our various asset classes. You will see that the largest impact was on our estimate of credit reserves required for our purchased performing loans. The Day one transition adjustment on these portfolios was approximately $8.3 million, the allowance for credit losses increasing and stockholders' equity decreasing, by that amount. We do not consider that this adjustment, which represents less than $0.02 of year-end 2019, GAAP book value, material to our financial position. For our purchased credit impaired loans, CECL adoption only affects balance sheet presentation with no impact on stockholders' equity. Subsequent to adoption, income recognition needs for this portfolio, will be determined using contractual cash flows, rather than on a loss-adjusted basis used previously. In addition, we will account for these loans going forward at an individual loan level, rather than on a pool basis. As we account for these lines at individual loan level rather on a pool basis, income recognition on any delinquent loans is subject to non-accrual accounting. As payments can be choppy on reperforming loans this may result in some additional variability in yields reported each quarter. Adoption of CECL has limited impact on our residential mortgage securities that we account for on, an available for sale basis. Importantly, we do not anticipate that CECL adoption in and of itself, will have any significant impact, on how we recognize income on our securities investments, including for our Legacy Non-Agency MBS. In addition one helpful change under CECL is that allowance for loan loss accounting replaces OTTI accounting. This means, that improvements in cash flow assessments that result in lower credit loss reserves immediately increase income rather than over time as an adjustment to yield. Loans measured at fair value through net income are not in the scope of the new standard, so there is no impact on these investments from adoption of CECL. Going forward, we expect that CECL will affect financial reporting for our credit-sensitive investments, primarily in two ways. Firstly, as we acquire loans, we are required to record an allowance for loan losses based on the life-of-loan expected cash flows, even on newly originated or loans that are performing at the time of purchase. Depending on the volume of loans acquired each quarter, the amount of allowance recorded may potentially impact our GAAP net income. And secondly, as we periodically update CECL credit loss estimates, based on changes in actual and/or forecasted portfolio performance the impact of loss – of loan loss reserves is reflected in GAAP net income for the period. We are currently considering how to reflect changes in CECL reserves in the determination of periodic core earnings and we expect to discuss that further in our Q1 2020 earnings presentation. Finally, CECL has no impact on the fair value determination of our loans and securities investments reported in our GAAP financial results. Accordingly, there is no impact on the measurement of Economic book value due to CECL adoption or ongoing CECL accounting for loan losses. And with that, I will turn the call over to Gudmundur Kristjansson, who will review more details of our investment activity and portfolio performance for the fourth quarter and the full year.

Thank you, Steve. Turning to page 12. The fourth quarter was an exceptional quarter for acquisitions as we purchased approximately $1.7 billion of assets in the quarter and grew our portfolio by approximately $500 million. Most of the acquisitions were concentrated in whole loans of which we acquired approximately $1.5 billion in the quarter. In total, 2019 was a great year for whole loans acquisitions as our average quarterly whole loans purchase volume was approximately $1.1 billion in 2019. This robust pace of acquisitions continues to be driven by our success in building out our non-QM fix and flip and SFR businesses over the last couple of years. We continue to experience elevated levels of early redemptions on NPL/RPL MBS and we opportunistically sold $170 million across CRT Non-Agency MBS and NPL/RPL MBS for a gain of approximately $12 million. Turning to page 13. Our investments team was very active in 2019 and successfully executed our strategy of expanding our holdings of non-QM fix and flip and SFR loans throughout the year. In total, we acquired approximately $5.3 billion of assets and grew our investments portfolio by $1 billion in 2019. Our residential whole loans and REO portfolio grew by 60% and we doubled our holdings of MSR-related assets in the year. We also strategically shrunk our Agency MBS portfolio by 40% in the year to reduce pre-payment risk and allocate capital to higher-yielding credit investments and sold approximately $250 million of CRT securities to take advantage of tighter spreads and mitigate prepayment risk on premium-priced bonds. Turning to page 14. Our strategic push into non-QM fix and flip and SFR loans continues to bear fruit. Our success in incorporating these loan products into our investment strategy remains the primary driver of portfolio growth. Since the third quarter of 2017, our whole loans portfolio has grown from about 19% of our investment portfolio to approximately 60% as of the end of the fourth quarter. The main reason for this growth has been the non-QM fix and flip and SFR loans, which have grown from zero in the third quarter of 2017 to over $5 billion at the end of the fourth quarter of 2019. In addition, non-QM fix and flip and SFR loans, now account for approximately 66% of the whole loans portfolio and approximately 40% of the total investment portfolio. We continue to benefit from our clear strategic planning and are extremely happy with the progress, we've made on incorporating these loan products into our investment strategy. Turning to page 15. MFA's investment strategy, which emphasizes credit risk over interest rate risk continues to deliver attractive yield and spread. Our credit-sensitive assets continue to benefit from positive credit fundamentals and lower prepayment sensitivity. Yield on interest-earning assets increased 34 basis points in the quarter to 5.66% and net interest rate spread increased 51 basis points in the quarter to 233 basis points. As Steve pointed out earlier, our asset yield was positively impacted by early redemptions on Legacy Non-Agency MBS. Excluding the impact of this, our asset yield would have declined eight basis points to 5.24%, and our interest rate spread would have increased nine basis points to 191 basis points. Short-term rates continued to fall in the fourth quarter of 2019 and the first quarter of 2020. In response to multiple Fed rate cuts lower expectations for interest rate and growth in general as well as recently due to concerns about the impact of the coronavirus on economic activity. One-month LIBOR declined by 74 basis points in 2019 and 12 basis points year-to-date in 2020. MFA was well placed to benefit from falling short-term rates in 2019 and continues to be well placed in 2020 as most of our asset financings are based on one or three months LIBOR rate and the percentage of repo that is hedged with pay-fixed swaps remains relatively low at about 35%. For context, our whole loan repo financing rate declined by approximately 78 basis points in 2019 and 67 basis points in the second half of 2019, as it grew from about a quarter of repo financing at the beginning of the year to about a half of repo financing for the end of 2019. Turning to page 16, where we show the yields, cost of funds, and spreads for our holdings, as well as the equity allocated to each asset class. Our largest asset class in terms of absolute dollar amount invested as well as equity allocated continues to be whole loans at carrying value with approximately 55% of MFA's total equity allocated to it. In addition, including whole loans at fair value, whole loans in general continue to represent the majority of our equity allocation at around two-thirds of MFA's total equity. Whole loans at carrying value yielded 5.27% in the quarter, while the leverage for this asset class increased in the quarter to 2.3 times debt-to-equity compared to 1.9 times debt-to-equity in the third quarter and 1.2 times debt-to-equity in the fourth quarter of last year, consistent with our communication throughout the year of increasing leverage on this asset class over time. Turning to page 17, where we take a look at MFA's interest rate sensitivity. Due to the growth in our portfolio in the fourth quarter and in particular our whole loans portfolio, our asset duration increased by 15 basis points, but remained relatively low at 175 basis points at the end of the quarter. Our hedged portfolio was relatively unchanged in the quarter. As a result, our net duration increased modestly in the quarter, but remained relatively low at 136 basis points at the end of the fourth quarter. Turning to page 18. As we continue to move through various interest rate cycles and experience large changes in interest rates over short periods of time, it is important to remember that MFA's strategy remains focused on maintaining a low and stable interest rate duration, while emphasizing credit-sensitive assets over interest rate-sensitive assets. Our strategy has consistently limited quarterly changes in the book value. As we have recently started disclosing Economic book value, we have added changes to economic book value to the graph on this page, which you can see as the dark gray bars on the left in the bar chart. Since 2014, the largest quarter-over-quarter decline in our book value or Economic book value has been 4%, with the average change in book value of less than 2%. By limiting book value fluctuations, we believe MFA will have the staying power to take advantage of new opportunities as they arise. And with that, I will turn the call over to Bryan, who will talk about our credit-sensitive assets in more detail.

Thank you, Gudmundur. Please turn to page 19. Mortgage credit continues to be supported by housing and economic fundamentals. Housing affordability has remained elevated as mortgage rates continue to be attractive for borrowers. Home price growth is accelerating again. The CoreLogic National Home Price Index was up 4% in November from a year ago. The unemployment rate seems to have plateaued at a historically low level of 3.6%. The supply of homes available continues its steady march downward. We are now seeing levels we haven't seen in over two decades. All these factors should continue to support stability and growth to home prices. The last reported 90-day mortgage delinquencies are just below 1% and we expect delinquencies could remain low as we believe low unemployment combined with prudent underwriting standards are good ingredients for loan performance. Turning to page 20. Our loan strategy had a record year of acquisitions. We acquired over $4.25 billion in loans in 2019 and $1.5 billion in loans in the fourth quarter consisting of approximately $1.2 billion of non-QM loans and $323 million of business purpose loans. We are developing and growing our existing relationships with origination partners and continue to add new partners. We didn't purchase many seasoned loans over the year, as we found spreads and absolute yield levels less attractive versus our new origination options and the performance of our seasoned loan portfolio has been very strong with the guidance of our asset management team. Turning to page 21. Our RPL portfolio continues to perform well and has remained stable. 87% of our portfolio remains less than 60 days delinquent. In addition, although 13% of the portfolio is 60 days delinquent or greater, almost 30% of these loans have been making payments over the last 12 months. Prepayment speeds remain elevated in the fourth quarter even with a moderate retracement in rates. With an amortized cost of $0.84 on the dollar prepayments are positive. Rates have rallied again so far in 2020. That combined with our borrowers gaining access to new financing options, as their credit improves should support prepayment rates. Turning to page 22. Our exceptional asset management team's oversight of servicing decisions and active management of the portfolio, have enhanced returns. The team has worked in concert with our servicing partners to more quickly get loans to reperform, as well as limit and reduce time lines to resolution. This slide shows outcomes for loans that were purchased prior to the year ended 2018 therefore, owned for more than one year. 36% of loans that were delinquent at purchase are now either performing or paid in full. 43% have either liquidated or are REO to be liquidated. 21% are still in nonperforming status. Our modifications have been effective as three quarters are either performing or have paid in full. We are pleased with these results as they continue to outperform our assumptions at the time of purchase. Turning to page 23. We had another successful quarter of growth to our non-QM portfolio. To date, we have acquired over $4.6 billion of UPB, including over $2.8 billion in 2019 and continue to work with our origination partners on strategic relationships. A variety of different loan types can be considered non-QM range from structural features, such as an interest-only period or a term greater than 30 years to the way income is documented such as the use of bank statements for self-employed borrowers or loans with higher debt-to-income ratios and so on. We believe the underwriting of these loans is prudent. The portfolio has a weighted average loan-to-value ratio of 67% and a FICO of around 716. The credit performance has performed as expected with approximately 1% of the loans 60 days delinquent or greater. And leverage is attainable through warehouse lines and securitization. Securitization execution has improved further through this year and we expect to utilize securitization as a financing option in 2020. Low interest rate environment, we currently live in yields on these assets trade in the low to mid-4% range and that we are able to achieve low double-digit ROEs with appropriate leverage. And now I'd like to turn the call back over to Gudmundur to walk you through our business purpose loans.

Thank you, Bryan. Turning to page 24. We've continued to actively acquire business purpose loans in the fourth quarter, as we purchased approximately $330 million in UPB and undrawn commitments in the quarter. Since we started acquiring business purpose loans at the end of 2017, we have purchased approximately 7,600 loans or $2.3 billion in UPB and undrawn commitments. We are excited about our progress and expect to continue to expand our acquisitions of business purpose loans in the future. At the end of the fourth quarter, we held approximately $1 billion UPB or fix and flip loans with additional $130 million of undrawn commitments. Our acquisition strategy remains unchanged with strong cooperation with origination and servicing partners with respect to underwriting standards and management of delinquent loans. Our target yield for this asset class is in the high 6% range. Our SFR portfolio grew by approximately $100 million in the quarter to $457 million at the end of the fourth quarter. Credit metrics and performance remains strong for our SFR portfolio and in line with expectations. Our target yield for this asset class is in the low to mid-5% range. With that, I will turn the call over to Craig for some final comments.

Thank you, Gudmundur. Please turn to page 25. In summary, we remain very active in the investment market. We purchased $1.5 billion of whole loan assets in the fourth quarter and $4.3 billion during the year, increasing our overall portfolio by $1 billion. This growth in our whole loan portfolio has resulted in materially higher interest income over the last year and capital structure optimization together with additional whole loan purchases should contribute to earnings as we move forward into 2020. The strategy that we deployed beginning in early 2017 has proven to be quite prescient and we believe that we are well positioned to continue to capitalize on these strategic initiatives. We are optimistic about our business and eager to move forward in 2020. This concludes our presentation. Operator, would you please open up the call for questions.

[Operator Instructions] And we'll go to the line of Doug Harter with Credit Suisse. Please go ahead.

Thanks. As we -- now that we're in 2020, can you talk about what's your expected pace for continuing to add kind of new production whole loans? Is the fourth -- was there something that caused the fourth quarter to accelerate? Or is this kind of accelerating trend something that we can expect to continue given the relationships you have?

Sure. We did have – Doug, we did have an exceptional quarter in the fourth quarter in acquisitions, but we do maintain to have robust sources of additional loans through now and our relationships with our origination partners sort of continue to provide us with ample amounts of loans. Will it necessarily always be that strong? I don't know. But we still believe that we have a really good runway for additional loans through 2020.

Great. And then I guess as you're looking forward, I guess to fund that growth, I mean I guess how are you thinking about further portfolio rotation of Agencies or Legacy Non-Agencies into these newer production assets versus looking to raise additional capital?

So -- thanks for the question Doug. So I think at least thus far with all the loan acquisitions that we did in 2019, we certainly didn't feel that we were capital constrained. In fact, I think you saw our leverage tick up a little bit. And we said consistently that we didn't feel that we had fully optimized the capital structure around the capital that we had already. In terms of rotating out of other asset classes, I think to the extent that there are opportunities to maybe rotate out of Agencies into credit assets, we've shown that we've done that. Unfortunately, because of the leverage on the agency portfolio, it doesn't really free up that much capital. So it's certainly a tool. But like I said, it doesn't free up all that much capital.

Got it. Thank you, Craig.

Sure.

And the next question in queue will come from the line of Henry Coffey with Wedbush. Please go ahead.

Yes. Good morning everyone. Great quarter.

Thanks, Henry.

So when you look at the investment horizon obviously the number one issue is economic and credit quality and it's -- we'll keep focusing on that. But what about the opportunity set? As it changes, is there more capital chasing QM or less in the fix and flip and rehab business? I know that's a big initiative of Fannie and Freddie, there abilities for you to enter into that side of the equation maybe with a different product? Or I also know Fannie said that they were going to be putting more assets into CRT. Will that improve the pricing of those assets? It's just -- it's a very efficient market but I'm wondering where the emerging opportunities are?

So it's hard to say Henry. I think as I said in my remarks, I think GSE reform we're not sure where it goes, but it's pretty hard to envision how GSE reform won't at least attempt to shrink the footprint of the GSEs. And to the extent that the footprint shrinks that could create investment opportunities for us. That said, I think we're seeing continued growth in the non-QM space. The fix and flip and single-family rental those are -- most of our origination partners continue to grow. There is a little bit more competition for those assets. I think there are more folks that pay attention to those worlds now than was the case a year or two ago. But again, I think that's where our strategic initiative with these originators that we started almost three years ago really puts us in a good position from a competitive standpoint.

Yes. I would just like to add too. I mean Craig has been very vocal about -- look there's all options are on the table for MFA in the resi space and so we continue to evaluate all kinds of opportunities. But we've been plenty busy with expanding the non-QM fix and flip and the SFR and we feel like there's a good runway for the foreseeable future.

What about CECL? CECL is in very -- many ways solving a problem that doesn't exist, but it does add volatility to your GAAP earnings. With your adjusted numbers, are you going to be able to give us a different metric or a different way of looking at those numbers? I know the SEC has been kind of strict about that. How are we going to be able to kind of look past CECL to get a sense of your earnings and dividend paying capacity?

Henry, sure, it's Steve. Thanks for the question. Yeah. As I sort of alluded to in my comments, we're still sort of evaluating the best way to reflect the economic impact of CECL in our results as we go forward. So it's a little hard to comment on that right now, because obviously we haven't -- we're only halfway through the first quarter and we don't really have a good sense of what our CECL numbers will shake out to be at the end of the first quarter. But you're right. We are thinking about how to reflect that in the Q1 core earnings metric, but you mentioned the SEC comments. We have seen those as well so we're trying to just evaluate what that means for us as well as we think about the topic as a whole.

Great. Thanks. Yeah.

And just to be clear, Henry, any CECL charge that runs through our income statement for purchased loans during that quarter where we setup a life-of-loan loss model, it's obviously a non-cash item and it's a non-tax item. So, I guess, as we think about it we're -- we understand that that's GAAP and that's how it works. But we also -- as I say we understand that it's a non-cash item and it's not a tax item either.

I mean, obviously, if you pay a $0.20 dividend and your economic book value is stable then that's a message to us that the overall earnings power of the business is enough to cover that $0.20 dividend.

Yeah. And I think we've been saying for probably two years that portfolio growth will drive earnings. And I think especially in 2019 you saw that, right? And we pointed out the interest income line was up sequentially each quarter. And it takes a while to acquire that many whole loans, but I think you're starting to see some real results from it.

Super. Thank you.

Sure, Henry.

And our next question here will come from the line of Eric Hagen with KBW. Please go ahead.

Hey, guys. Thanks. Good morning. One question on your capital structure. You guys have your preferred equity in your baby bond MFO up for redemption as soon as next month. I'm just curious what the rate incentive would need to be for you guys to call and refinance that capital and make it accretive to shareholders?

So thanks for the question Eric. Obviously those markets have been active away from us so far this year. And suffice to say we're on top of what's going on and we're aware of those opportunities.

Okay. Okay. And how should we think about the servicing expenses for the non-QM portfolio? I really appreciate the additional detail on delinquency rates in that portfolio, but -- and I think we agree that the credit performance in that segment of your portfolio will probably continue to be very stable, but is the servicing costs going to be proportional to the level of delinquency in those loans? Just curious how that line item shakes out.

Hey, Eric. Thanks for the question. It really -- as it relates to the servicing expense for the non-QM loans and similarly some of the new origination loan, it really depends how the assets are acquired. If the assets are acquired where we own the servicing that expense is going to run through. If we acquire them net, it's just that the net coupon runs through the income side and you don't really see -- there really is no servicing expense that runs through. So it's -- really it just depends on how the asset is acquired.

Okay. But sort of the nature of my question though is really did servicing costs increase in proportion to the level of delinquency for those loans?

So not really. Because -- so if we're buying – especially, if we're buying the loan servicing is retained. So there is a strip that -- a fixed strip that the servicer retains. And whether it's more expensive for them to service if a loan goes delinquent that sort of -- that's their problem it's not -- that's baked into the cost. So it doesn't increase -- incrementally increase our servicing expense.

Got it. Okay. Thanks for that color. That's really helpful. Thank you.

Yeah. Sure.

And we're at the line of Stephen Laws with Raymond James. Please go ahead.

Hi. Good morning.

Hey, good morning.

I wanted to follow-up a little bit on Doug's questions about the originators and just the pipeline of the non-QM loan investments going forward. Do you currently -- I think you said five partners. Are you looking to grow that? And then what percentage of their origination do you guys currently acquire? Is it 100%? Or is it something less than that that you'd like to increase? Kind of how do you look at what your capacity is in the pipeline that you're choosing from?

Sure. Thanks for the question. So as far as the five originators we have a lot more than five originators that we buy loans from. It's five originators that we've made capital investments in. As far as what percentage we buy, we don't buy 100% of anybody's production. I would say in most cases, we probably don't even buy 50% of the production from most of the originators. And I think that's as it should be. I'm not sure we'd ever want to be at 100%. So it's hard to say exactly what the pipeline is, although, I will say that for the most part all of the originators that we have relationships with and across all those product types it's probably give or take 20 or so, they're all continuing to grow. So, I think, we're pretty optimistic, although, again, it's not something that we can forecast with any precision, yeah.

And on the financing side, as you think about using securitization and the benefits that -- I know there's a number of benefits that provides, but what would the cost savings be as you think about where similar deals or pricing with similar collateral to what would have in the securitization? Kind of how do you think about potential -- the leverage that would be in the structure as well as potential cost savings versus the current financing that you're using?

Sure. As it relates to the leverage, it just gives us the confidence to sort of take-up the leverage a bit when we have sort of non-recourse, non mark-to-market financing versus having loans on repo. So where -- we sort of said over the past, however, many quarters and years, we sort of considered ourselves to be a bit under-levered. So that really -- when you're under-levered you're not really too concerned about where your financing sources are coming from because you have incredible amounts of liquidity. As you continue to grow the portfolio and use warehouse lines that's when you think about okay maybe I should diversify the financing sources into a securitization-like format. In terms of the cost savings there -- as LIBOR has come down dramatically they're not as much as they were previously, but they're still meaningful to be in the order of 10 basis points to 20 basis points to 30 basis points given where spreads are sort of moving around. But we do see it as a good option for us as we continue to grow the portfolio.

Yes. And how much of it would do you guys intend to retain? The bottom 5% or a bigger piece of the stack? Or how do you think about what you'll retain there?

So it would really depend on pricing, right? So I think as we look down the stack and consider how deeply to sell down we're looking at the yields on those. And at some point it starts to look like an expensive liability. And therefore it looks like an attractive asset. So that's just sort of how we think about that. There's no magic number. If I had to guess I would guess that we'll retain certainly more than 5% for sure.

Okay. Great. I appreciate the color on that. And then lastly on ATM you said you put it -- it's back in play. Have you guys utilized it year-to-date? If so how much? And how should we think about using the ATM versus reallocating capital? I know that was covered a little bit but there's minimal equity in the agency portfolio. But how do we think about the ATM going forward?

Yes. Thanks, Steve. I mean we did put the ATM back in place in the third quarter of the year. And we had it let lapse for a number of years. So we really -- did it really just to add another tool to our toolbox. We did some disclosure in our third quarter 10-Q of usage and that will be carried forward to our 10-K which we expect to file a little later today. It's been very minimal to date in terms of the usage. I think it was roughly -- I'm just going from memory here. I think it's roughly about one million shares. So very minimal usage at this point. But suffice to say, it's a tool in the toolbox. And we could -- we'd use that when it makes sense to do that going forward.

Right. Appreciate the time.

Thanks, Stephen.

And we'll go to the next question. It will come from Rick Shane with JPMorgan. Please go ahead.

Hey, guys. Thanks for taking my question this morning. Look, I think, slide 13 actually really validates Craig's description is prescient in terms of the shift over the last several years when we look at investment flows. I am curious as the portfolio shifts how you think about the return characteristics of the portfolio that you're building and the trade-offs in terms of risk whether it is credit or leverage? Will you be able to achieve the same returns and maintain the same sort of risk controls?

So thanks for the question, Rick. I mean I think the short answer is, yes. I think as we look at these new investments -- and clearly these are mortgage credit investments. We're -- and it's different from the credit that we've had exposure to in our past right? Because when we started buying Legacy Non-Agency back in 2008 and 2009 these are loans that have been made years before and we're buying them at a discount and it was a very different world. These are newly originated loans. And so I think we're very cognizant of LTVs for instance because at the end of the day that's your biggest protection. And I think -- I'm happy to say if you look across those portfolios the LTVs are pretty low right? They're in the mid to high 60s or so. We've been around long enough to look at this in a somewhat jaded fashion. And over time will LTVs increase? I can't tell you but we'll certainly have our eyes wide open if we do see origination start to change. And as far as leverage I think the leverage -- I don't think the leverage will change all that materially. I think if you go back years our leverage was probably between 3 times and 4 times and that's probably appropriate given this asset class. If it's repo or warehouse line borrowing for loans and it's a 20% haircut then mathematically we couldn't be more than 4 times levered. We might get a little bit more through securitization. And as Bryan said maybe securitization makes us more comfortable. But I don't see it really deviating much from sort of low 3s to mid-3s.

Got it. And look when we look at the trends, obviously, the whole loan portfolio is going to continue to grow. We would expect the MSR portfolio continue to grow the Legacy and RPL/NPL MBS likely to shrink. I guess the one question I would have is do you consider Agency MBS to be still an opportunistic opportunity that's terrible. But you get you understand what I'm saying. Is that something that we will see flex based upon spreads widening or contracting and what your opportunities are in the other core portfolios?

I think, as Gudmundur said earlier, I think, we say, everything is on the table every day. So, we can't possibly forecast, what will happen with spreads in all these various asset classes, over the next year or so. That said, I think, we've made a clear choice that we see better risk return characteristics on the mortgage credit side, at least in the past. But that could change. And we're not allergic to agency mortgages by any means.

Thank you very much.

Thanks,

And we'll go to line of Steven Delaney with JMP securities. Please go ahead.

Thank you for the question and congratulations on a strong year, especially in your whole loans. Steve, I'll -- Stephen I'll touch on this, but I'd like to ask if you could give us some color on, what you're seeing in the universe of private mortgage originators of NQM and BPL products out there. Are you seeing down in circo? I mean, it's a very entrepreneurial industry. But are you seeing more new shops coming up, that are picking up the phone and offering product? Or is it still a relatively limited number of potential counterparties? Thanks.

So, I guess, if you take non-QM, I would say, there are -- I wouldn't say there are a lot of new entrants in the non-QM space. That said there easily are dozens of them out there.

That's helpful to know dozens. That frames it, Craig. Thank you.

But I don't -- we don't see a lot of new entrants there. On the business purpose particularly the fix and flip side, there's probably -- there could be hundreds of them out there. And a lot of them are very small. And so it's -- there are some bigger players, in that space, but there are a lot of players. It's a very fragmented space.

That's helpful. And I assume you've bought loans from more originators than you have in terms of just your strategic partners the five. Can you give us any sense from how many different originators whether it's NQM or BPL you may have worked with over the last several years, just a rough idea.

Sure. So I said earlier, that the -- so we've made capital investments in five originators, but we transact with about 20.

Okay.

So it's not 100. We're not a conduit.

Understand.

And we know all the entities very well. But the number is probably somewhere in the vicinity of 20.

That's helpful. And we recently saw or read of the sale of one of the early and large originators of NQMs to a New York credit fund. I'm just curious if -- I don't know whether you did business with that counterparty. But I know they were fairly significant. Do you have any idea whether -- despite the change of ownership, whether they will still be selling product to other investors? Or do you think it's, the buyer of that entity is just buying it to control 100% of the flow? And obviously, I'm asking this on a no-name basis.

Sure. Sure. So we're obviously familiar with that. I think, it's probably too soon to tell what happens to that production, but we'll see.

Okay. And you've been consistent. But I'm just going to go ahead and ask. Is it still -- in terms of the way you see the market and the opportunity, do you still see that you'll be able to accomplish your objectives without having to go into the origination business yourself? Is that still the -- sort of the outlook?

Yeah. I mean, we've looked at those companies for sale. And I'll never say that we'll never do that. But I think -- so far, I think, we're pretty happy with the strategies, we've deployed where we've made some -- we have some strategic partnerships. We've made some capital investments. But we haven't bought a company, but again there could be an opportunity some time in the future we'll never say never but I think we're pretty happy with the strategy that we've deployed so far.

Right and Steve, remember, if we buy 100% of an entity as Craig mentioned prior, we contract with 20 different partners. So if we buy one of them one of the other 19 all of a sudden thinks of us…

Absolutely.

…So we're very cognizant of maintaining good relationships with all of our partners.

Well, what you're doing is working. So keep up the good work. And thanks for the comments.

Thanks, Steve.

And for -- currently no further questions in queue.

Okay. I want to thank everyone for joining us. And we'll speak to you again, next quarter.

Thank you. And ladies and gentlemen, this conference will be available for replay after 1 o'clock today, and running through May 21 at midnight. You can access the AT&T replay system at any time by dialing 1-866-207-1041 and entering the access code 1598787. International parties may dial 402-970-0847. Those number again, 1-866-207-1041 or 402-970-0847, with the access code 1598787. That does conclude our conference today. Thanks for your participation and for using AT&T teleconference. You may now disconnect.