使用警語:中文譯文來源為 Google 翻譯,僅供參考,實際內容請以英文原文為主
Operator
Ladies and gentlemen, thank you for standing by, and welcome to the Great Ajax Corp. Q2 2020 Earnings Call. (Operator Instructions) I would now like to hand the conference over to your speaker today, Mr. Lawrence Mendelsohn, CEO. Thank you. Please go ahead, sir.
Lawrence A. Mendelsohn - Chairman & CEO
Thank you very much. Thank you, everybody, for joining us on our second quarter of 2020 earnings call. Before we get started on the presentation, I want to just have you take a look at Q2 -- or at Page 2 for the forward-looking statement disclosure.
And with that, we can get moving to Page 3, our business overview. Q2 2020 was a positive quarter in many ways. After navigating through March and April, we closed the second tranche of joint venture investment in loans that was purchased in a prefunded securitization structure that we created in March of 2020.
We raised $125 million in net proceeds of perpetual preferred shares. We paid down significant amount repurchase facility debt in Q2 as a result of loan and mortgage bond cash flow. And beginning in late Q2 2020, we've seen our cost of funds start coming down dramatically versus the average Q2 cost of funds.
One thing I do want to add, volatile environments and declining economic times like we've seen in the last few months really show how important and strategic having our directly aligned operating and loan servicing platform can be, coupled with our value investing mentality. You can see this through the continued performance of our loan portfolio and the net asset value relative to book value.
On Page 3, our managers' strength in analyzing loan characteristics and market metrics for reperformance probabilities and pathways and our managers' ability to source mortgage loans enables us to acquire loans that, we believe, have a material probability of long-term continuing reperformance. We've acquired loans in 302 different transactions since 2014.
Additionally, we believe having an affiliated servicer provides a strategic advantage in nonperforming and nonregular paying loan resolution processes and timelines. In today's volatile environment, having our portfolio teams and analytics group with the manager working closely with the servicer is essential to maximize reperformance probabilities on a loan-by-loan basis.
The analytics and sourcing of the manager and the effectiveness of the servicer also enables us to broaden our investment reach through joint ventures with third-party institutional investors.
Our June 30, 2020, corporate leverage ratio decreased from 3.2x at March 31 to 2.2x at June 30. We use primarily moderate leverage, especially for our sector and primarily non-mark-to-market leverage. Mark-to-market financing represents less than 25% of our financing and is approximately 90% backed by our Class A1 securities from our joint ventures.
On Page 4, we'll talk in-depth about the second quarter. Net interest income from loans and securities, including a $4.3 million partial reversal of COVID-19-related loan extension and credit losses, was approximately $15 million. For the second quarter, we had a lower average balance of mortgage loans due to prepayment and principal amortization but a higher average balance of investments in joint ventures that are on balance sheet as securities and beneficial interests.
Our joint venture loan acquisition that closed in April was on balance sheet for approximately 9 weeks, although its related interest expense was for the entire quarter since we prefunded the acquisition through securitization in the first quarter of 2020.
A GAAP item to keep in mind, though, is that interest income from our portion of joint ventures shows up in income from securities, not interest income from loans. For these joint venture interests, servicing fees for securities are paid out of the securities waterfall. So our interest income from joint venture securities is net of servicing fees, unlike interest income from mortgage loans, which is gross of servicing fees.
As a result, since our joint venture investments have been growing faster than our direct loan investments, particularly in the second quarter, GAAP interest income will grow slower than if we directly purchased loans by the amount of the servicing fees, and GAAP servicing expense will decrease by the offsetting amount.
An important part of discussing interest income is the payment performance of our loan portfolio. At December 31, 2019, approximately 76% of our loan portfolio by principal balance made at least 12 of the last 12 payments as compared to only 13% at the time we purchased the loans. At March 31, this number was 74%. And at June 30, approximately 72.6% of our loan portfolio by UPB made at least 12 of the last 12 payments.
In our Q1 2020 investor call, we mentioned that we expected COVID-19-related economic environment would negatively impact the percentage of 12 of 12 borrowers in our portfolio in the second quarter. Thus far, the impact on payment performance has been less than expected, although the effect on interest income is still difficult to quantify and forecast.
While regular paying loans produce higher total cash flows over the life of the loans on average, they can extend duration. And because we purchase loans at discounts, this reduces percentage yield on the loan portfolio and interest income.
However, regular paying loans generally increase our net asset value, they also enable financing at a lower cost of funds and provide regular cash flow. Loans that are not regular monthly pay status tend to have shorter duration. However, we expect that this duration reduction will be less than typical due to market conditions and the impact of COVID-19.
As I mentioned earlier, most of our loans were purchased as nonregular paying loans. And the borrowers, our servicer and portfolio team have worked together over time to reestablish these loans as regularly paying. We also expect that given the low mortgage rate environment and the stability of housing prices so far that prepayments will likely continue to increase on both regularly paying and nonregularly paying loans over time.
Our cost of funds in the second quarter was higher than in the first quarter by 11 basis points, which equates to approximately $300,000 of additional interest expense for the quarter, even though our average outstanding asset level debt was $25 million less. This was primarily due to temporarily higher rates on certain of our mark-to-market securities repurchase facilities caused by market disruption, which were partially offset by lower cost of funds for our mortgage loan repurchase facilities.
Net income was $6.2 million or $0.27 per share after subtracting out $1.84 million of preferred dividends and $735,000 of income attributable to noncontrolling interests.
A couple of other things to note. We recorded $700,000 of flow-through income from the June 30, 2020, mark-to-market of our managers and servicers ownership of Great Ajax common shares versus the March 31 mark-to-market. Our manager and servicer combined own approximately 1.1 million shares of our common stock, and we own 19.8% of our manager and 8% of our servicer. Also, we expensed approximately $1.4 million relating to the GAAP required accrual of the warrant put rights from our second quarter issuances of preferred stock and warrants.
Book value rebounded from $14.37 a share at March 31 to $15.20 at June 30. This is still lower than the $15.80 per share at year-end 2019. The increase from Q1 primarily reflects a $23 million mark-to-market adjustment to our debt securities as generally determined by marks provided by our financing counterparties after a $28 million negative mark-to-market adjustment in Q1. Almost all of our debt securities that are subject to mark-to-market repurchase agreements are Class A1 securities from our joint ventures.
Taxable income was negative $0.09 per share. The 2 leading drivers of lower taxable income in the second quarter were fewer foreclosures and more REO sales as well as increasing loan ownership through joint ventures. Taxable income declined in the second quarter primarily as a result of lower foreclosure gains versus prior quarters, while the level of tax losses from REO sales remain elevated.
For tax purposes, we recognize a gain at the foreclosure date in an amount equal to the difference between our tax basis in the loan and the fair value before expenses of the property. As the tax accounting at foreclosure does not take into account liquidation expenses, a portion of this gain is typically reversed when you sell the REO. As a result of COVID-19 effect on REO and strong pay history of our loans, we have very few new foreclosures, which reduces tax gains, and more REO sales, which increases tax losses.
Additionally, our joint ventures are accounted for as single loan pools with the purchase discount of those pools recognized over the remaining weighted average life. Under the tax law, the pricing speed and yield of the loan pool is required to remain constant over the life of the pool, and loan level gains are not recognized. Accordingly, the historical gains previously recognized at the loan level for a foreclosure or modification event are no longer recognized, and all income is effectively recognized as OID.
Cash at June 30. We had approximately $163 million of cash at -- and for the second quarter, we had an average daily balance of $125 million. Our surplus cash currently tempers earnings a bit, but this should reverse as we get the cash invested over time. In addition to cash, we also have approximately $65 million of unencumbered BBB-, BB- and single B-rated securities from our securitizations and approximately $25 million of unencumbered Class A2 and B1 securities.
As I mentioned earlier on the call, over 72% of our portfolio by UPB made at least 12 of their last 12 payments compared to only 13% at the time of loan acquisition. I will discuss the importance of this in greater detail when we get to Page 9 on the call.
If we flip to Page 5. It's pretty clear that we continue to be primarily RPL-driven, with RPLs representing approximately 97% of our loan purchases. We strive for positive payment migration of purchased RPLs.
And on Page 6, you'll see that we continue to buy and own lower LTV loans. Our overall RPL purchase price is approximately 54.4% of the current property value and 87.7% of unpaid principal balance.
And on Page 7, our purchased nonperforming loans, NPLs, have been declining relative to the total loan portfolio. For NPLs on our balance sheet, our overall purchase price is 73% of UPB and 52% of the current property value.
We've not changed our target markets in Q2. California continues to represent the largest segments of our loan portfolio. Our California mortgage loans are primarily in Los Angeles, Orange and San Diego counties. We have seen consistent payment and performance patterns from loans in these markets. We've also seen consistent prepayment patterns even in recent months. Thus far, the impact on our California loans from COVID-19 and its related economic effects has been relatively muted.
We removed Las Vegas as a target market during Q1. Mortgages in Las Vegas are currently a small percentage of our portfolio, unlike 5 years ago. Our analytics suggest that COVID-19 will have a material economic impact on Las Vegas, given its tourism focus and the economic multiplier effect, although this will be partially offset by tax rate-related transitions that we've seen from Southern California.
We are keeping a close eye on Houston, as the combination of COVID-19 and oil industry struggles is having a material impact. It has not spilled into the single-family homes market nearly as much as the apartment market thus far, but we have trimmed back Houston targets. Since we only added Houston to our target portfolio locations in the second half of 2019, it's a very small percentage of our portfolio.
We had been seeing material negative effects from the new tax law SALT provisions in New York City metro area, in suburban New Jersey and Southern Connecticut home values and home sale liquidity. We have seen a quick positive turn in liquidity in these suburban locations as a result of COVID-19, as New York City apartment dwellers look for suburban residences. It is too early to tell, however, whether this is a short-term phenomenon or a longer-term change in lifestyle as a result of COVID-19.
Related to this, we have also seen demand for homes and home rentals surge in parts of Florida, but we've not seen the same for condominiums in Florida.
If we turn to Page 9, portfolio migration. At June 30, approximately 72.6% of our loan portfolio made at least 12 of the last 12 payments, including 65.6% of our portfolio that made at least 24 of the last 24 payments. Again, this compares to approximately 13% at the time of purchase.
As a result of the economic impact of COVID-19, we've seen a small decline in these numbers. During Q1, we had a 2% decline in loans as a percentage of our portfolio that are 12 of 12 payments or better. And in Q2, the decline was 1.4%. We expect that this number will continue to be affected, but it is too early to determine the materiality.
Since we generally purchase lower LTV loans and our purchase price to property value averages in the high 50s percentage, we would expect the principal impact of a decline in regular paying loans to be duration extension, as borrowers with loans that are paying but are not 12 months contractually current will have more difficulty refinancing.
Since we buy loans at discounts to the face amount of UPB, duration extension reduces the speed in which we receive the discount on the affected loans and, therefore, the interest yield. COVID-19 duration extension can also affect the yield on true nonperforming loans, as extended resolution timelines can lead to more property tax, insurance and repair expenses over that extended timeline.
Since we purchased most of our loans when they were less than 12 for 12 payment history, our servicer has worked with most of our borrowers over time. While it's too soon to understand the full effects of COVID-19 on home prices and mortgage loan performance, so far, the impact has been less than anticipated for us.
12 for 12 loans in today's market trade at materially higher prices than pre-COVID-19, primarily related to the rated securitization cost of financing now. As a result, we believe our portfolio and related implied corporate net asset value are materially higher than GAAP book value, which represent our loans at amortized cost.
On Page 10, post-Q2, we closed on approximately $46 million principal balance of residential first mortgage loans. The purchase price was approximately 88% of UPB and 66% on value of the underlying properties. We also have another $11 million of principal balance of loans under contract to purchase subject to due diligence.
On July 30, we priced Ajax Mortgage Loan Trust 2020-B with approximately $97 million of AAA-rated senior securities and $17.3 million of single A-rated securities with respect to $156 million UPB of loans. The AAA- and single A-rated bonds represent 73.2% of UPB and approximately 82% of cost basis. As a result, for the $156.5 million of underlying loans, we have approximately 4.5x leverage at a weighted average interest cost of 1.88% that is non-mark-to-market and nonrecourse.
We declared a cash dividend of $0.17 a share to be paid August 31 to holders of record on August 14. As a result of COVID-19 uncertainty and the related economic impact, taxable income is harder to forecast. We do, however, anticipate that our 2020-B securitization will trigger significant taxable income.
Our current joint venture partners continue to work with us seeking investment opportunities, and we anticipate additional material joint venture transactions this year.
If we turn to Page 11, financial metrics, just a few things to note. Average loan yields, excluding reserve recapture, declined by about 70 basis points. This is primarily as a result of duration extension from both paying loans and nonpaying loans. As discussed earlier, the principal effects we are seeing from COVID-19 impact is duration extension.
Also remember that yield on debt securities and beneficial interest is net of servicing fees and yield on loans is gross of servicing fees, as I discussed earlier. Debt securities and beneficial interest is how our interest in our JVs are presented under GAAP. As our JVs increase, as they did in Q1 and Q2 relative to loans, the GAAP reporting will show lower yields by the amount of the servicing fees.
Our leverage continues to be quite low, especially for companies in our sector. We ended Q2 with asset level debt of 1.9x. While our asset level debt cost of funds was higher in Q versus Q1 due to market disruption in the securities repurchase agreement financing market, the cost of our asset level debt has begun to decline considerably beginning in late Q2 and even more so in Q3. And you can certainly see that from our 2020-B securitization.
As we get our surplus cash invested as well, we should see material increases in interest income and net interest income as well. And with -- that sums up my remarks. If anybody has any questions, we are happy to answer.
Operator
(Operator Instructions) Your first question is from Tim Hayes from B. Riley.
Timothy Paul Hayes - Analyst
My first question here, just on the reversal this quarter. Can you just help us understand what drove that? Was it as simple as just better collection rates than you expected? Or were there inputs like home values or interest rates that were the bigger drivers of the reversal? Just trying to understand how volatile this could be going forward and how likely the reversal could actually be reversed going forward.
Lawrence A. Mendelsohn - Chairman & CEO
It was primarily collection rates. And cash flows were significantly more than we had estimated when we, in late Q1, early Q2, built our kind of COVID expectation model. That being said, we expect it to be volatile in the world of CECL.
So we took a reversal of a reserve that we thought was the proper reversal. The cash flows continued to be significantly better than we expected from COVID. But -- even through July. That being said, there's no way to know kind of how this all turns out as we come into the school season in the fourth quarter.
Timothy Paul Hayes - Analyst
And can you maybe share why you think cash collection has been -- and credit performance has been better than expected? I mean does it just have to do with, I don't know, the demographics of your loans? Or is there anything more kind of from a macro perspective that you think is driving better credit performance? Just any color you could share on that would be helpful.
Lawrence A. Mendelsohn - Chairman & CEO
So far from discussions with borrowers and looking at performance, one of the most important things is our loans tend to be lower LTV and have more absolute dollars of equity. And as a result, on average, the borrowers are -- the borrower's personal momentum has been derailed a little less than lower-value properties.
Our properties tend to be worth -- we talk about deciles. Our properties tend to be in deciles 4.5 through 7, kind of primarily for predictability and liquidity. And in this environment, we've seen that be an advantage in terms of absolute dollars of equity.
I also think, to a smaller extent, much smaller extent, stimulus money has been helpful, although I think it's less available or had less of an impact on our portfolio than in mortgage land in general.
Timothy Paul Hayes - Analyst
Okay. Got it. And then what is the current forbearance rate on the portfolio today? And how has that trended so far in the third quarter?
Lawrence A. Mendelsohn - Chairman & CEO
Sure. We've had -- for our portfolio, we've had about 15% of borrowers call, but about 80% of the 15% continued to make payments regularly. And about 3%, which is why you've seen our 12 of 12s go from 76% to 72.4%, have stopped making payments.
Timothy Paul Hayes - Analyst
Okay. 3%, got it. And then I don't know if you can disclose this or if there is a view internally. I understand this is a very opaque environment and a lot yet to be seen. But what loss rate scenarios are you using internally for -- whether it's just kind of cash flow or credit analysis that you're modeling internally?
Lawrence A. Mendelsohn - Chairman & CEO
It's loan by loan by loan. We build our models loan by loan by loan in terms of expected losses and cash flows rather than kind of a CPR/CDR mentality. We go loan by loan by loan. And based on loan characteristics, we have different probability paths of paying or not paying, and of not paying what potential outcomes are and, of paying what potential outcomes are.
And that's -- so we base everything off of those individual loan-by-loan probability paths. So there isn't really what I would call a global prepayment or default rate. And then based on probability paths, the absolute -- the property values versus the underlying loan amounts plus bridges, if any, we determine losses, if any, and timelines.
Timothy Paul Hayes - Analyst
Okay. So you -- I guess the point -- what I'm trying to get at here is, your stock is obviously implying a massive amount of losses in your portfolio. And I'm just wondering, when you run your kind of probability scenarios and your loss rate scenarios that if you think the market got it right or if you think that they are severely over-discounting the stock based on what you expect and just kind of what's driving the --.
Lawrence A. Mendelsohn - Chairman & CEO
So our portfolio is kind of unusual in that if we had 100% defaults, the yield would probably increase, but NAV would decrease. So -- because 100% defaults with shortened duration, which would mean yields would go up because of the low LTVs. But we're much more focused on NAV and long-term cash flow. So we would never want that to happen. We would want the opposite to happen.
We'd rather have 100% payment, even though it would reduce interest income in any given period, but it would increase NAV dramatically and reduce financing costs super dramatically. I mean we've learned that if 100% of our loan was clean pay 12 of 12, we could finance the entire portfolio 1.88% and at 4.5x leverage with a cost basis on the loans in the 80s. And we've seen -- we've seen 12 of 12 prepaid loans recently sell for basically 3% yields.
Timothy Paul Hayes - Analyst
Right. And I guess just on your comment there, though, about the inverse relationship with yields and NAV there. So maybe this is more reflecting -- or maybe this is what the stock is more reflecting is taxable income and the dividend here.
The dividend was well covered by GAAP earnings this quarter, but taxable income, negative, and last quarter was very low. So you're paying out more than you need to. And just based on kind of -- I think you're probably looking back (multiple speakers).
Lawrence A. Mendelsohn - Chairman & CEO
But we also are paying out because we expect we'll have to.
Timothy Paul Hayes - Analyst
Right. Right. Okay. Well, I guess my question was just, again, like touching on the view of taxable income in the second half of the year and how you're thinking about where the dividend is set here in that context. That's it. Okay. All right. Fair enough. I'm going to hop back in the queue.
Lawrence A. Mendelsohn - Chairman & CEO
Yes. One thing I would add, Tim, is tax and GAAP, when you have no assets remaining, tax and GAAP, over the years, have to be the same. So tax eventually has to approach where GAAP has been. So we would anticipate that some of these securitization structures trigger loan sales as well -- trigger loan taxable income, prepayment increases will trigger loan taxable income. And the market value of our clean pay loans could very well trigger some taxable income at some point as well.
Operator
The next question is from Stephen Laws of Raymond James.
Stephen Albert Laws - Research Analyst
I hate to beat up on the dividend, but I want to pursue -- Tim's line of questioning. For taxable income, I've got it negative $0.04 first half. I think it was a $0.05 and a negative $0.09. And I just kind of want to understand your and the Board's thoughts on establishing the dividend, where it is, and whether or not you're going to continue to review it quarterly.
I understand you've got the cash flow to overpay taxable income. And I think in your prepared remarks, you talked about the 2020-B securitization will increase TI. But you also said you expect more JVs this year, which was one of the reasons, more loans owned in JVs that you had cited a lower tax in 2Q.
So trying to, one, reconcile the right level, but also, you're raising money with the preferreds to deploy because you like the new investment environment. So would it make sense to reduce what you're paying out to have more cash that you can retain and put into new investments that you're not having to pay such a high fee for the -- like you are on the private placements?
Lawrence A. Mendelsohn - Chairman & CEO
So I could argue it both ways, and I think some of our institutional investors could argue it both ways as well from discussions over the years with them. We think there's a lot of opportunity, particularly for value investors in this environment, who can have a longer timeline mentality in terms of buying things and monetizing.
But on the flip side, from an NAV perspective, we have a material built-in gain, I would -- for lack of a better descriptive term. And if COVID-19 were to become very bad and we had a lot of foreclosures, our taxable income would soar. If we sold the loan, our taxable income would soar. And some of our rated securitization structures now will trigger some taxable income as well, a material amount of taxable income.
As a result, we're -- our Board has taken the position where, let's keep it, the dividend, where it is for now until we see how COVID's impacts are on loan performance and whether that triggers more foreclosures or more defaults and, hence, more taxable income over time, or whether it keeps loans paying economically and our loans are -- continue to be less impacted, as we would expect, as we saw.
But we're evaluating kind of tax planning strategies to accelerate taxable income to some extent because tax/GAAP is so different. And what we don't want to do is wake up 3 years from now and have an enormous amount of tax relative to GAAP because there's some giant catch-up that happens. So our Board is evaluating each quarter. And right now, they think that the $0.17 dividend is pretty descriptive of what we anticipate taxable income would cause distributions to be through the remainder of this year.
Stephen Albert Laws - Research Analyst
Very helpful. And touching on the taxable income but not quite for the dividend. But with the JV and the sales taking place there, what operations are taking place in a taxable REIT subsidiary, where you could technically, I guess, retain the earnings if you wanted? Or my understanding is you can dividend them up to the REIT level and distribute out. But how much income do you have taking place at the TRS versus at the REIT?
Lawrence A. Mendelsohn - Chairman & CEO
A fair amount takes place at the TRS. We do -- our multi-class cash flowing securities generate some income. But we have income that happens at the TRS from real estate and our interest in the manager --.
Unidentified Company Representative
And servicer.
Lawrence A. Mendelsohn - Chairman & CEO
And our interest in the servicer as well. So -- and the manager, we own 20% of the manager on a 0 basis. So we pick up 20% of the manager's income through a TRS. And we pick up 8% of the servicer's income through the TRS.
Stephen Albert Laws - Research Analyst
Okay. That's helpful. Last question for me is on the expense side for 2 of them. Real estate operating --.
Lawrence A. Mendelsohn - Chairman & CEO
And we have some securities in our TRS also, but not a material amount relative to our total securities.
Stephen Albert Laws - Research Analyst
Okay. And on the expense side, looks like real estate operating expenses, much lower than we've seen. On the other side, other expenses, a little bit higher than we've seen historically. Any color what's going on in those 2 line items? I haven't read the full filings, so I apologize if I missed it out there somewhere.
Lawrence A. Mendelsohn - Chairman & CEO
Yes. Real estate impairments are down considerably in real estate, some of which happened -- some of which is related to real estate selling quickly as a function of COVID-19, okay, number one. Also, we only have about $8 million of real estate owned remaining in our portfolio because so much has sold and so many loans are paying, very little new real estate is being created.
And on the other expense side, you see an increase of other expense by about $1.2 million or something like that. That is the accrual of the warrant put-related expense.
Operator
(Operator Instructions) We have a question from Kevin Barker of Piper Sandler.
Kevin James Barker - MD & Senior Research Analyst
Can you hear me okay?
Lawrence A. Mendelsohn - Chairman & CEO
Yes. Yes.
Kevin James Barker - MD & Senior Research Analyst
So on the acquisition front, I mean, COVID has thrown a curveball on -- with foreclosure moratoriums and everything like that. Could you help us think about the timing of when you're going to see the best opportunities to make acquisitions over the next several quarters, just given the foreclosure moratoriums, changes in home prices, different appetites for real estate?
I mean there's a lot of things that are really changing right now in the real estate market. And I just would love to hear your thoughts on when do you think that ramp is going to occur.
Lawrence A. Mendelsohn - Chairman & CEO
Sure. Well, in Q3 so far, we've acquired about $45 million pretty cheaply in many different transactions. And we're finding there's a lot of small originators who have bigger problems than big originators have with scratch-and-dent loans. The other thing -- so from a value investment perspective, it's -- those purchases are -- it's still 2014 and 2015.
The -- in terms of real estate markets and kind of the opportunity set, the real estate market that we're seeing and from our discussions with borrowers and people on the ground, and also we have an interest in Gaea property REIT that we started, and we're seeing it from our -- the apartment as well, is that each market is much more different than it was 5, 6 months ago and in a magnified way.
The -- for example, in Florida, we're seeing significant demand for homes, especially from people in the Northeast, on the East Coast of Florida, and the Midwest, on the West Coast of Florida, looking for homes but not condominiums.
So we've seen dramatic liquidity both for purchasing single-family homes, people purchasing -- people wanting to sell single-family homes like us as REO. Or -- but even for rentals, we've seen dramatic liquidity increases, just like you've seen in the New York metropolitan area and probably in the Philly suburbs as well, where you are.
In the New York metropolitan area, houses that might have been on the market for 2 years now have 6 offers in a weekend. And -- where in the city, rents are down and significantly less demand for high-rise buildings.
So these are phenomena that have come dramatically around in the last 4 months. And it's hard to tell whether they're permanent, temporary or somewhere in between, or semipermanent in that some piece of it will continue for tax reasons or for some other reason and not COVID over some period of time.
Now the other thing that I will say is that people underestimate the power and impact of mortgage rates going from 4% to 3%. That on a $400,000 house, a 1% change in mortgage rate saves the buyer $4,000.
If you think that houses are basically a 5-cap product, that increases the value of that house by $80,000 just from the 1% change in mortgage rate because someone can buy it for $80,000 more and have the exact same monthly payment. And we've seen that effect also as a result of a decline in interest rates, very obviously economic and COVID-related. And how long that lasts is also a question.
So as a result, we're extremely focused on buying low LTV loans in this environment because we don't know whether the effect is permanent or temporary. And we hate losing money. So as a result, we are being pretty careful in an extreme kind of what I'll call value investing mentality right now.
And while we have a lot of liquidity and it's a drag on earnings, we don't know whether there'll be a round 2 or not that you're going to want to have a lot of liquidity for because there's going to be some great opportunities even better than we're seeing now. So we're -- right now, we're buying things that we see that are cheap rather than looking at momentum is the way I would describe it.
Kevin James Barker - MD & Senior Research Analyst
I mean what -- do you see that as an opportunity to maybe recycle your existing portfolio for gains and it's --?
Lawrence A. Mendelsohn - Chairman & CEO
Yes. Yes. There's no question about it that given our clean pay loans, to the extent we see significant opportunity to -- on the buy side, that increases yield given where clean pay loans trade in today's environment, we may very well recycle some of our portfolio.
Kevin James Barker - MD & Senior Research Analyst
Okay. So would this be a defining moment where you start to really -- we're going to see a significant amount of gains being generated, a higher amount of --?
Lawrence A. Mendelsohn - Chairman & CEO
Well, you just have to be careful -- because remember, REIT rules, there's only so much you can sell in a year as a percentage of your tax basis, no more than 20% in 1 year and a 10% rolling 3-year average. So there are limitations as to what you can sell.
Kevin James Barker - MD & Senior Research Analyst
Okay. So it seems like you're going to have elevated cash for the foreseeable future as you become more opportunistic down the road.
Lawrence A. Mendelsohn - Chairman & CEO
Yes. Yes. To quote one of our large shareholders, this is the time to have your Warren Buffett/Berkshire Hathaway hat on and think about what you can buy really cheap and monetize over 3 or 4 years versus what you can try to ride the wave on.
Operator
There are no other questions in queue.
Lawrence A. Mendelsohn - Chairman & CEO
If there's no other questions, thank you, everybody, for joining our second quarter investor call. We're always happy to talk about our business. And we appreciate you joining us, and we appreciate the questions, and look forward to talking to you between now and our next call. And I hope everybody stays healthy and safe.
Operator
Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.