New York Mortgage Trust Inc (NYMT) 2019 Q3 法說會逐字稿

Good morning, ladies and gentlemen, and thank you for standing by.

Welcome to the New York Mortgage Trust Third Quarter 2019 Results Conference Call.

(Operator Instructions) This conference is being recorded on Wednesday, November 6, 2019.

A press release with New York Mortgage Trust third quarter 2019 results was released yesterday.

The press release is available on the company's website at www.nymtrust.com.

Additionally, we are hosting a live webcast of today's call, which you can access in the Events & Presentations section of the company's website.

At the time, management would like me to inform you that certain statements made during the conference call, which are not historical, may be deemed forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995.

Although New York Mortgage Trust believes the expectations reflected in any forward-looking statements are based on reasonable assumptions, it can give no assurance that its expectations will be attained.

Factors and risks that could cause actual results to differ materially from expectations are detailed in yesterday's press release and from time-to-time in the company's filings with the Securities and Exchange Commission.

Now at this time, I would like to introduce Steve Mumma, Chairman and Chief Executive Officer.

Steve, please go ahead.

Thank you, operator.

Good morning, everyone, and thank you for being on the call.

Jason Serrano, our President, will also be speaking on this call this morning.

The company continued to deliver solid results.

We generated GAAP earnings per share of $0.15 and comprehensive earnings per share of $0.20 for the quarter.

Our book value per common share was $5.77 at September 30, an increase of $0.02 from June 30 and resulted in a total economic return of 3.8% for the quarter.

The company raised $311 million in common equity through 2 overnight offerings during the quarter.

In October, the company also completed a preferred equity offering, raising an additional $167 million, bringing our total stockholder capitalization to over $2 billion.

The company has generated an annualized total economic return of 17% through the 9 months ended September 30, while maintaining our dividend at $0.20 per share for the quarter, its 11th quarter in a row.

Our investment team remained very active, sourcing and funding approximately $400 million in credit-sensitive assets, bringing our total investment portfolio to $4.5 billion at September 30.

Our pipeline of investments for the fourth quarter is strong, as we expect to close over $275 million in credit investments in both our multifamily and residential assets.

I'd now like to have Jason to speak to the -- about market conditions and some thoughts on our core portfolio strategy.

Good morning.

Global and U.S. equity markets have made modest gains during the third quarter of 2019 despite volatility, largely driven by investor uncertainty regarding global trade restrictions and alongside other geopolitical concerns, which is becoming increasingly harder to predict.

While economic data remains mixed, data released for the third quarter suggests that the U.S. economy has continued to expand in line with the previous quarters at approximately 3%.

While GDP growth and the labor market data indicate modest economic expansion, consumer and business confidence indices have weakened.

Reflecting our expectations made throughout the year, recent survey data continues to indicate business activity slowing.

However, with respect to the U.S. housing market, arguably the 2 most important factors are unemployment and interest rates.

Both these data points continue to provide a strong tailwind to the U.S. housing.

The U.S. labor market expanded during the third quarter of 2019, bringing a lower unemployment rate print of 3.5%.

And the 30-year fixed-rate mortgage is now over 75 basis points lower than at the year -- at the beginning of the year when the Federal Reserve started lowering rates, 25 basis points in the last quarter, which is the third time this year.

Strong employment and lower rates, improved the cost of homeownership.

In fact, the household debt service ratio related to a residential mortgage is now at the lowest point in 40 years.

These points have mostly offset changes to the 2017 tax code, which disincentivize owner-occupied housing.

As such, property price forecasts project depreciation by 3% annually over the next 2 years, equal to the current national growth.

On the supply side, only 4 months of housing on the market for sale and near low vacancies of -- record vacancies at multifamily demonstrates a U.S. housing market that is in great shape.

With this backdrop, we are pleased with the results of our rotation from agency to credit.

One of the areas we have focused throughout the year is in single-family portfolio of residential credit securitization bonds.

This sector represents 14% of total assets at quarter end.

While we did not -- while we did monetize depreciation experienced in selected securities in the quarter, on the balance, we expect price appreciation to continue, particularly in esoteric markets where securitization in asset classes that were created after the crisis.

We see strong technicals that will likely growing tighter.

Total new bond issuance in the single-family securitization market is approximately $100 billion year-to-date.

The issuance exceeded market supply forecasts for the year.

Despite this increase, runoff in the sector from bond refinancing and organic asset paydowns is likely to bring flat net overall supply of residential credit bonds in the year.

Against a backdrop of solid underlying fundamentals and little growth of new issue residential credit bonds, we expect a demand-supply imbalance to persist for better selling opportunities.

With that said, we have developed channels to source bonds outside the market's general auctions.

With our familiarity in -- as an active investor in assets that are similar to the underlying pools that are securitized, we can move quickly to efficiently price securitized bond risk without the need of a rating or performing data from the securitization market.

This has helped us create proprietary channel of investment securities that are off market where we compete on process and not just price.

Within our residential loan portfolio, we have selectively aggregated a portfolio of $1.3 billion or 30% of assets over the last 12 months.

We have been more active in what we categorize in our public filings as distressed loan paper.

As mentioned on prior calls, the profile of distressed loans we are buying is a selective mix of borrowers that are in a payment gray area.

These borrowers are generally either a few months delinquent or just a few months current on their mortgage loan.

After the credit crisis, where over 7 million borrowers were processed through government modification programs, a permanent solution to a loan restructure was a missed opportunity for many loans today.

Banks and the GSEs continue to derisk from loan servicing.

Servicing delinquent or recently delinquent loans is not a core business function for these entities and are finding channels to derisk with large-scale portfolio sales.

Given the frequency of multibillion-dollar portfolio sales throughout the year, Big Data and analytics is key to the investment strategy, we seek certain borrower profiles where we have experienced favorable results to tailored servicing outreach programs.

The higher quality velocity after -- the higher payment velocity after loan boarding improved our interest income in the quarter.

We expect to build on higher interest income related to this strategy in subsequent quarters as we exit the J-curve of new portfolio acquisitions made earlier in the year.

Given our activity in the securitized debt market, leading to a consistent view throughout 2019 of tighter bond spreads, we have been patient with the timing of [what] to lock in term financing our loan positions from our bank warehouse funding.

Term financing cost, which is typically a fixed rate in this market, is now 100 basis points lower than at the start of the year.

We would look to lock in these lower term funding rates with certain loan profiles where we feel execution is favorable in the near term.

Now with single-family mortgage rates approximately 77 basis points lower than at the start of the year, borrower refinancings or CPRs are up in almost all residential mortgage loan sectors.

That benefits our loan portfolio because of the discount to par paid to acquire the asset.

However, a different part of our business has benefited more directly.

Due to the rapid rate decline, many originators were caught off-balance with the spike in refinance volume.

Volume spikes create a strain on origination operations because it's more difficult to scale up personnel than down for volume inflections.

Given the myriad of compliance and underwriting guideline measures that were installed with Dodd-Frank, we generally see a higher volume origination process errors created.

In our scratch and dent loan business, we help originators face near-term liquidity -- they're facing near-term liquidity constraints by buying these loans intended to be securitized with the GSEs or non-agency channels.

But it can't be sold because some miss in the underwriting process.

As a result, we have a nice buildup in our year-end investment pipelines of these assets.

Because we can move quickly to price the assets, the underwriting issues, we have experience with buying over -- loans from 100 different counterparts in this market, we're able to drive volumes in this niche subsector.

The value proposition of our business of buying loans or pools of loans at substantial discounts where the intention was to sell immediately or flip at a premium to the securitization channel is excellent.

Now switching over to the multifamily side of the business where approximately $1.5 billion or 33% of our total assets are invested as of September 30.

We continue to see strong fundamentals with vacancy rates remaining near historic lows.

Starts on multifamily homes containing 5 or more average -- seasonally adjusted rate of 367,000 during the 9 months ended September 30, 2019, which is at a similar pace to last year.

Overall, headline supply of unit deliveries remained strong.

However, we continue to see misallocation of the supply.

With stubbornly high labor and material costs at relatively the same level across different regions in the United States, builders focus on where rents are the highest to maximize the project's economics.

With this constraint, we see elevated construction in prime gateway markets like L.A., San Francisco, New York and Miami.

These developments have access to relatively cheap funding costs because of the high -- because the size of the familiarity of the market, especially from overseas financing counterparties.

However, in these same markets, we are observing negative domestic state-to-state migration.

Residents looking for cheaper cost of living and lower taxes are starting to have negative impact on supply and demand technicals.

While reported examples include residents in L.A., who are finding better value in Dallas or New York City residents that are moving to Florida like -- in markets like Tampa Bay.

It's not hard to envision the California resident population of 40 million to decline in the coming decade.

This would be the first time California had population decline even going back to the mid-1800s.

Other markets with high new construction such as New York is already seeing population decline.

What's even more troubling for these markets is that a supermajority of the construction is in the luxury sector.

According to Yardi data, last year, a staggering 87% of all new construction of at least 50-unit apartments was dedicated to the luxury high-end segment.

This focus provides uneven growth opportunities in south, southeast, where migration is positive and supply is lacking, especially for workforce housing, which is a particular focus for us.

Due to these trends, we see significant growth opportunities for our business.

As an example, in the fourth quarter, the company expected to fund its fourth Freddie Mac K-Series first loss investment for the year totaling $57 million.

Recently, FHFA provided updates to their multifamily loan purchase cap.

Through the 5-quarter period of Q4 2019 to Q4 2020, Fannie and Freddie will each be allowed to aggregate $100 billion of new loan production.

With the MBA's expectation of $390 billion of multifamily lending next year, the GSEs will continue to be a predominant player in this market.

Not only will robust securitization of K-Series deals continue to be sold by Freddie Mac, we're also expecting the GSEs to look for new ways to pare down their accumulated portfolio holdings in other ways.

In fact, last month, Fannie Mae began this effort with its first-ever risk transfer multifamily deal, which we participated in.

We have been a consistent partner of the enterprises and look forward to expanding that relationship in new ways in the new year.

Since our first purchase in the K-Series investment in 2011, we have evaluated the financials and underwrote over 1,200 U.S. multi-tenant properties across the United States.

Out of these, we have conducted on-site inspections and review the capability of the project manager on over 650 properties, 200 in the last 1.5 years alone.

The amount of on-the-ground market knowledge that we have gained through our participation in the first loss holder of these deals allows us to stay ahead of the headline risk.

The competitive edge is enormous and provides for unique off-market opportunities.

With this knowledge, we have built strong relationships with sponsors of multifamily properties across the U.S. As an example, our direct mezzanine loan program to multifamily properties continues to add substantial value to our earnings, with 15% of our capital allocated to this strategy as of September 30.

In this space, we relied on our proprietary relationships, particularly in the south, southeast sectors of the United States.

In fact, we have focused on the region -- this region of the U.S. for years now, which has consistently provided attractive risk-adjusted returns in the double digits.

We consistently win deals largely due to our efficient process and a solid track record of meeting our commitments.

Lastly, as I stated earlier, we continue to rotate out of the agency trade as net interest margins compressed throughout the year in this sector.

Without dramatically increasing leverage, we find it difficult to consistently provide attractive returns.

Negative convexity risk in this market is very difficult to contain through hedges.

Interest rate volatility tends to increase the cost of hedging, which is why we had 7.7% of capital exposed to this sector and declining.

The expectation from here should be incrementally lower exposure quarter-over-quarter, at least until we see rates flatten out.

With the recent cap raises, my goal today was to provide you with a full explanation of our focus in the credit markets and how the business transitioned over the years take advantage of feedback loops generated between our core strategies.

We are excited about our ability to deliver strong earnings and under low levels of leverage utilization of 1.5x today.

At this time, I'll pass it back to Steve for closing comments.

Thanks, Jason.

And I'd like to go through a little bit more detail of our income statement and what we've done in the third quarter.

We had $34.8 million in GAAP net income and $45.7 million in comprehensive net income.

The company generated a net interest income of $32 million and had a portfolio net interest margin of 240 basis points for the quarter.

Our earnings strategy over the last several years does not rely solely on net interest margin or as many defined core earnings to cover our dividend, but on the total contribution for all income sources, including fees gain on sale and unrealized activities.

Our net margin typically represents about 60% to 65% of our earnings in any given quarter.

But by generating earnings away from that margin, we take some of the leverage pressure off of our balance sheet as our leverage is less than 2x and is currently 1.5x, which we believe reduces volatility in our earnings quarter-over-quarter and has allowed us to maintain the stable book value, enabling us to deliver a $0.20 dividend for 11 quarters in a row.

Our average earning assets totaled $3.9 billion for the quarter, an increase of almost $360 million from the previous quarter, bringing the total increase for the year of $1.2 billion or 43% since the beginning of 2019.

We expect our investment portfolio to be approximately $5 billion by year-end, with average earning assets of $4.5 billion for the fourth quarter as our investment pipeline continues to remain very strong.

Our investment portfolio totaled $4.5 billion at September 30, including $956 million in agencies -- Agency RMBS securities, with $142 million or 7.7% of our total capital, as Jason alluded to.

Our focus remains on credit-sensitive investments, which are priced generally at a discount or near par to minimize our exposure to interest rate convexity risk, mainly prepayments, which is a significant component to this agency return.

We have $2 billion in residential credit investments, including subperforming and reperforming loans as well as non-agency securities backed by varying types of residential credit loans.

These investments are currently funded with repo lines, but we continue to evaluate both the rated and unrated markets for possible securitizations.

We anticipate our first securitization will be completed in the first quarter of 2020.

We have $1.5 billion of multifamily investments or 37% of our invested capital.

In the fourth quarter, the company expects to fund $56 million in our fourth Freddie Mac first loss security of the year.

We believe the fundamentals for renting remains very strong and will continue in the future as the dynamics of homeownership versus renting evolve.

Our net interest income totaled $21.4 million for the quarter.

Included in our press release for the first time are several tables by income category within noninterest income to try to better explain these components as there are many.

We had net realized gains for the quarter of $6.1 million, primarily from the sale of certain CMBS securities.

We had net unrealized gains of $11.1 million for the quarter, really comprised of 2 significant components: One, a $13.3 million loss from interest rate hedges, which was offset by $24.4 million in unrealized gains from both our multifamily and residential loan portfolios.

We had other income of $3.9 million, largely comprised of income from our multifamily direct lending investments.

But for accounting purposes, we -- they do not qualify as loan accounting, and therefore, the income generated from those assets must be characterized in noninterest income, even though the actual risks are identical to the loans that we're including in that margin.

Our general and administrative expenses were $8.3 million for the quarter, down $1.5 million from the quarter ending June 30, 2019.

The decrease was due to 2 main components: The termination of our last external management agreement in the second quarter, which resulted in a reduction of management fees of $575,000 as well as a $700,000 decrease in expenses related to an annual equity Board compensation award that has been on the second quarter.

On a normalized basis going forward, we would expect the expenses to run at approximately $8.5 million per quarter.

As we head into year-end, we believe the company has never been better positioned.

Our market cap exceeds $2 billion for the first time in company history.

We have a strong and growing pipeline of targeted credit investments and a team of professionals focused on our mission, to deliver stable book values over longer periods of time while maintaining a stable dividend to our shareholders.

We appreciate your continued support and look forward to speaking about our annual and fourth quarter results in February 2020.

Our 10-Q will be filed on or about Friday, November 8 of this week with the SEC and will be available on our website thereafter.

Operator, if you could please open up the call for questions for Jason and myself?

Thank you.

(Operator Instructions) And our first question will come from the line of Stephen Laws with Raymond James.

I appreciate the comments in your prepared remarks, Steve and Jason.

Can you talk about -- you've raised a good bit of capital in the last few months.

Talk about incremental returns you're seeing?

And I guess coupled with that, I know you guys did touch on this in your remarks, but what are you seeing the most attractive between the multifamily and resi?

I know you mentioned that the agency is likely to run down here.

But where are you seeing the most attractive opportunities today?

And other than agency, are there any asset classes that you guys are really just avoiding at this point?

Yes.

I think, look, when we think about raising capital and we look at where we're going to deploy it, Stephen, over the last, really, 3 years, it's almost fallen 50-50 into residential credit and multifamily.

And right now, we see equal opportunities in both classes.

But there's a varying difference between those classes, right?

In the multifamily side, the investment involves much more credit analysis on a specific asset, either in the direct lending or a combination of several larger assets in the Freddie K first loss securitizations.

Those we're going to fund with a lot less leverage than we would typically in the residential portfolio.

And on the residential side, we continue to focus on credit.

And maybe Jason can speak further on the residential side.

Yes.

So on that side, we are seeing multibillion-dollar portfolio sales on a quarterly basis from both the GSEs and large banks.

Our goal of those portfolios is to find a portion of those portfolios where it meets our guidelines for acquisition, where we have servicing strategies and data that's helping us understand the ability for that bar to be predisposed to a positive outcome.

So we particularly will partner on transactions where we can take a subsector of a larger pool and acquire it for our purposes.

So we still see equity-type returns in that space, which is the subperforming loan space.

As the bids on cleaner paper continue to ratchet up with lower rates, that's providing a nice tailwind for gains in our portfolios.

In the residential space, we still don't see a large market opportunity in the non-QM trade.

We avail ourselves to all originators in the market as we're not competing directly against any one of them.

The supply in that space continues to underwhelm.

A lot of the refinance supply obviously at lower rates.

It has compressed some of the equity returns.

But overall, aggregating a larger portfolio for a securitization is taking -- would take a longer period of time, which exposes you to more rate risk that we would like.

The only area of the market, I'd say, that we kind of have -- one area that we definitely have pushed back quite a bit is in fix and flip.

While we have aggregated portfolios in that space selectively, our concern there is that the refinance volume of borrowers that are going through fix and flip rather than selling their home after a year.

And again, most stay back.

This a market where borrowers are basically taking bridge loans for a year to renovate a property and sell it at profit.

In that space, the renovations to a sale are now renovations through a refinance then to a sale.

So when you start seeing a pickup of refinances in this market versus a sale of a home, you know that the original project parameters were missed, something was missing in the project for it to not be sold within that year period.

So that will -- we believe that may add risk to that sector over time and cause a pickup in defaults.

So that's an area that we're a little bit more concerned about, just simply looking at the time lines of the sales that are occurring by these local developers that are taking these houses through a sale.

So overall, as I said in my comments, looking at the SPL market, looking at scratch and dent, given the higher volumes of origination activity and more mistakes that are made has continued providing us with attractive equity returns.

And Jason, following up on the portfolio sale commentary, how should we think about opportunities in the fourth quarter?

As the -- with multiple holidays in the back half of the quarter, do things slow down or with year-end are larger financial institutions more incentivized to try and sell these underperforming portfolios off their balance sheet?

How do we think about the investment opportunity and seasonality in the fourth quarter?

Yes.

I mean, every year, we've been doing this for 7, 8 years in this space in the loan space.

That I've been doing it.

There always is window dressing towards the end of the year for some institutions that provide some opportunities, which requires our traders and analysts to be at the office over the new year period, but -- for closing.

So that is more opportunistic in nature and generally at bigger discounts.

So that opportunity will avail itself towards the end of the year.

It almost always happens.

But if you look back the last 2 years, we've had quite a bit of volatility in December.

I think the market, and talking to different participants out there, I think we should expect a slowdown of trading activity, simply because of that volatility.

Last year was obviously very tough for the market.

And obviously, there was a significant bounce back in the first quarter.

But at this point, there has been profits made on the residential credit sector.

And it would seem more likely that everybody would try to print their -- stabilize their book by not entering into large-scale trades in the fourth quarter.

With that said, we obviously are -- we'll be open for business and we'll likely see some opportunities.

But again, it will be more opportunistic in nature.

We're not expecting any significant large portfolio sales.

Those sales are happening in this month.

And Steve, if you think about the amount of capital that we've raised over the last 12 months and you think about our earning assets relative to the ending quarter balance because of some of the assets that we're investing in, especially residential loans, there's probably a 60-day lag between entering into a letter -- some kind of purchase agreement in the actual settlement of the loan.

So we sort of have a rolling natural 400-ish type pipeline of investments that are going to fund in the fourth quarter.

And so a lot of the activity that we would typically do in the fourth quarter, you're not going to really see it until the first quarter of 2020, especially as it relates to a lot of the residential investing.

But I think we feel pretty good about from a funding standpoint, where we're going to end up at the end of the year.

And as Jason said, I think there just was a large Fannie Mae loan sale this week.

There will probably another one next week from another seller.

And then as we go into the end of the fourth quarter, you will see sporadic selling that we think we can take advantage of.

Great.

And my last question, flipping to the financing side of the equation.

It looks like the financing cost for the resi credit declined, the multifamily credit actually went up a few basis points.

Can you talk about what drove that?

And with LIBOR moving lower and obviously the forward LIBOR curve now pretty flat, can you talk about what you expect to see occur on the financing cost as we look forward?

Yes.

And really, it was really the financing cost increase in the multifamily was really more of a function of increasing the leverage there and extending the term of some of the financing, Steve.

So to the extent that we're able to get 12- and 18-month repos on -- that's a 2-year repos on some of our Freddie K securities.

And so we just felt like just to be prudent to take some of the volatility out of the funding side is to extend some of those maturities, and that's where some of that cost increase went to.

There was a little disruption in LIBOR pricing as we went into the quarter end.

So as we started to think about refinancing our repo book in September, we start pushing some of the repos out over year-end just to avoid the year-end possible price issues.

And so that's probably why you saw a little tweak up in cost, but no significant increase in cost.

And in fact, as we look at our warehouse lines and our loans, we're in the process of finalizing a new agreement, which will lower our cost in the residential loans.

And so we think funding still seems to be very advantageous.

It's just a matter of transitioning stuff from 30-day exposure to a little bit longer, less risk exposure.

I'll add.

From the funding side on -- from warehouse lines, and we are seeing spreads tighten, roughly 20 basis points to 30 basis points, which we will be able to benefit from as we renew these lines.

On the LIBOR side, we saw 1 month LIBOR decline about 38 basis points over the year, which has saved us on an annualized basis roughly $8 million per year in total financing cost.

So that obviously is -- the financing side of the equation is very helpful to overall business in generating return for the assets we own.

There is an interplay that occurs between that and the assets.

The assets -- when the investors are looking at a targeted teens returns, if you look at the equation of lower funding LIBOR cost, and that does tend to increase the price of the next portfolio sale.

So while there is benefit in the lower LIBOR, there is an offset of slightly higher asset cost.

So we have to obviously look at the markets to where there's been a delay in that activity and that reflected in the current pricing in the market.

But we're very happy about our ability to obviously gain just the savings here and also our timing of when we would look to securitize as funding costs have come down 100 basis points in the securitization market.

So that has been a good result for us and something we probably look to take advantage of in the next quarters.

And our next question will come from line of Eric Hagen with KBW.

The loan pipeline with the GSEs, it sounds like conforming originations that you talked through in the prepared remarks.

Can you walk through that again?

I just want to understand why you guys were able to buy loans at a discount to par when they would be able to sell in the specified pool market for quite a hefty premium?

Yes.

So what we're talking about here are loans that failed some underwriting guidelines at the point of origination.

Simply giving a borrower notice of 60 days prior to closing loan of what is -- of what interest cost he would incur and other items like that or a borrower that may have taken out an auto loan in the period of which he was underwritten to the period of which it was closed, so therefore, the DTI, debt-to-income ratio changed slightly.

So these are things that we see that has persisted in this market for -- since the beginning of the -- really, the agency market.

And these defaults, when they come up, basically are -- have an issue in that the loan that is on a warehouse line that was intended to be sold to the agencies, generally, now it is disqualified for being on that line and the originator has a certain period of time of removing those assets off of the warehouse line.

And originators are basically close to 20x levered entities.

They have 95% kind of -- advance rates plus on the loans that they originate.

So in that case, there's a near-term liquidity issue that, that originator would face, especially with respect to nonbank lenders.

In that case, we have deployed strategies here where we work with originators, look at the issues that they have uncovered on their loans as they attempted to sell into the pipelines, and we would look to buy those loans at discounts, given the time line required to get those loans off and the infraction that was caused.

Now we're typically looking for technical infractions here, fraud and other items like that, that do come into play.

Those are not loans we look to buy.

And this is more of a kind of fire sale opportunity for us to acquire assets.

Now in this market, with origination volumes increasing and dealing with the supply and the origination framework, and it's harder to add employees to deal with the origination versus the other way around.

So in this case, the originators are kind of strapped for the operational side of the equation that they have -- when you have increase in origination and more supply going through the same pipeline without increasing your total production capacity, that you do find more issues.

So our team has been busy evaluating those, and we've seen almost a doubling of our pipelines in that space simply because of the refinance activity that's occurred in this market, given lower rates.

Very interesting.

No, I think that just supports the niche opportunity one gets in New York Mortgage Trust.

That's very interesting.

And then on the distressed loan segment, I'm just trying to gauge how fast the runoff is.

I don't think it's right to maybe phrase the question in terms of a prepayment speed or a CPR in the portfolio, but how much capital is being returned back to you in any given period, just in that kind of segment of the portfolio?

Yes.

I think if you -- historically, Eric, I mean, the rotation, really is a couple of combinations, right?

And your point is well taken about prepayment.

I mean, we really -- there is a lesser amount that comes from direct prepayment.

It's really either we throw them into a permanent type vehicle with securitization or we sell the loans outright.

Because we've been in a -- so we really -- we've accumulated a significant amount of loans over the last 12 months, we transitioned them to a new servicer.

So what we're trying to do is bring these borrowers to a different place in their credit cycle that improves the value before we think about selling them.

And so I think what you'll start to see as we go into 2020, increased velocity in selling some of these loans or putting them into permanent vehicles for financing to try to monetize that part of the credit improvement.

We have not been significant active sellers of loans in this year, primarily because we've been in the building portfolio phase.

And we still think that we can -- the loans that we're keeping, we think that we can continue to add value to them, and that's really why we haven't sold any.

Or sold very little.

(Operator Instructions) Our next question will come from the line of Matthew Howlett with Nomura.

Another strong quarter.

I just wanted to talk about the upcoming securitization, where -- what's the opportunity there?

What do you think -- you said 100 basis points in -- the market has come in 100 basis points.

Are we talking about 3% term type securitization that would take your overall costs down?

Could you just maybe just go into a little more detail what you're expecting early next year?

Yes.

As it relates to the securitization market, obviously, you're getting term financing, which is going to have slightly wider total cost of debt versus short-term repo funding.

However, it is termed out to the risk of any kind of mark-to-market or any other events like that is no longer an issue given the term financing.

So when you look at securitizations, you look at it as a function of the overall life of that financing, not just day 1. So given the 100 basis point decline, it's become more competitive or at least more competitive or closer to kind of repo financing terms, given the flat curve or inverted curve earlier.

So it's providing us with opportunity to basically term out with slightly increases on cost of financing.

We talked about a securitization.

We -- New York Mortgage Trust, we will be looking to be a frequent issuer in this market over time given our aggregation of loan portfolios that we will be receiving and earning the NIM off of those securitizations is what our business plan is.

There's a gestation period of when that happens, given the payment profiles of borrowers that were buying that would have to -- that we're looking to make more stable.

When you get to a point where the borrower has paid 12 consecutive months, it starts opening up the channel for really efficient financing where you can get a rated securitization done.

And that is our goal.

In the interim process, there is -- before that event happens, as you buy new loan pools, there are esoteric unrated securitizations that can be completed on our portfolios, where financing cost has come down, as I said earlier, about close to 100 basis points.

And in that case, were -- we'd be taking subsectors of our portfolio and moving them to a more of a term funding type of facility where we see the esoteric market pricing fairly the risk that is being assessed there.

I mean, again, we are buyers of the securitization debt.

We've analyzed and unrate the loans in those portfolios almost daily as new deals come down and looking at our current portfolio.

So we're constantly monitoring where the inflection point is for us to go into a securitization and which type of assets should go in securitization based on our pricing views on the securitized debt.

We have financing facilities in place that are -- that will be evergreen rolling facilities.

So we will have long notice periods of when the repos would -- for repo rolling, which has helped us alleviate the concern of any kind of cash needs in the near term.

But the securitization side of the equation is becoming more attractive given these esoteric markets are becoming more of a normalized market where empirical data has been provided to the bondholders, and they're getting more comfortable with respect to the risk that they're taking in the underlying deals, given that these are generally new securitizations that have been produced in this market.

Got it.

So when I think about NYMT's leverage profile going forward, these securitizations should enhance really the economic leverage, given you're going to be terming out stuff and freeing up capital for reinvestment.

Is that the way to sort of think about the impact to earnings going forward?

Yes, I think if you -- like we talk about callable leverage and total leverage of the company.

And so what you're doing is transferring some of your callable leverage risk to permanent financing, which -- it may increase our leverage in the securitization, but it doesn't increase our risk with the leverage.

And so that's really what we're evaluating.

And to Jason's point, again, trying to factor in how much money we can monetize against the loans and what's the cost of doing that relative to our shorter-term financing options.

Got it.

Okay.

And then flipping to Fannie.

Fannie, it's -- I just want to go back to your -- those comments you made on Fannie selling first loss risk with Freddie.

We know Freddie has that well-developed K-Series.

Talk about Fannie's program.

I know they've sold the DUS bonds, the AAAs, and it now looks like they're going into, what, sort of copying Freddie.

Can you look -- kind of go over the opportunity there?

Is it the same type of return profile?

Yes.

Look, the Fannie program is different than the Freddie in the sense that the Freddie is the true securitization where they sell the bottom layer risk out to the private marketplace.

In the DUS program, what Fannie is doing currently is they're accumulating a series of risk that they hold because of the DUS program.

And when you think about the DUS sharing -- loss-sharing program, it's more of an insurance policy with the originator rights.

So the originator is sharing some of the risk with Fannie Mae.

Fannie Mae is selling their portion of their risk, not 100% of the risk on a particular loan.

So unlike a Freddie K first loss bond where we actually have a chance to oversee the collateral in a workout situation, as a holder of the first loss piece, in the Fannie Mae, it's more like a residential transfer program that both agencies have, where you're just buying a bond based on an expected yield and you don't really have any ability to go in and deal with the collateral.

And so your -- all your analysis is upfront.

And so you're looking at that and you're trying to figure out the yield you're entering into.

So it was a very liquid market.

It was very well bid, the first transaction.

In some cases, the bonds were 40x oversubscribed.

So I think that they are taking advantage of an undersupplied market in this type of product.

And so it's really these first couple of transactions are going to be fact-finding in terms of trying to determine what is the right yield.

So I think we entered into the first trade; one, we wanted to make sure that we were -- we know the Freddie program very well.

We want to get an understanding by getting involved.

But it's really -- at this point, it's more of a wait and see, and see what really long-term opportunities are going to be.

We think at a minimum, in the short term, it's a trading opportunity.

And over time, we'll have to see how it plays out from an actual credit analysis possibility.

Great.

Well, look forward -- you guys have done -- been sort of the first mover with the Freddie Ks so look forward to hearing developments on Fannie.

Appreciate the comments.

Thanks, Matt.

And our next question will come from the line of Christopher Nolan with Ladenburg Thalmann.

Did I hear you correctly that operating expenses in the fourth quarter should be $8 million to $9 million?

Or did I mishear that?

That's right.

About $8.5 million, we think the run rate will be, Chris, something around that.

And Steve, do you think that's going to be the case at least in the first half of '20?

Yes.

I mean, look, I don't -- the company has experienced tremendous growth, both in capital portfolio size and human capital, which is obviously the large component of G&A.

I think the significant hires are behind us, and we can continue to grow the portfolio.

I mean, at the margin, if we continue to grow the company larger, will there be slight increases in expenses?

Possibly.

But I think the run rate is pretty solid now.

I mean, we've done a lot of growth in 2019.

Thank you very much.

And I'm showing no further questions in the queue at this time.

So now it is my pleasure to hand the conference back over to Mr. Steve Mumma, Chairman and Chief Executive Officer, for any closing comments and remarks.

Please proceed.

Thanks.

I just want to thank everybody for being on the call, the continued support of the company.

And we think we're in a great position right now as a company in terms of opportunities, and we will continue to push the company forward as we've done in 2019.

We look forward to going through our fourth quarter update and year-end in February of next year.

Thanks, everyone.

Ladies and gentlemen, thank you for your participation on today's conference.

This does conclude our program, and we may all disconnect.

Everybody, have a wonderful day.