PennyMac Financial Services Inc
NYSE:PFSI
Earnings Call Analysis
Q4-2023 Analysis
PennyMac Financial Services Inc
PennyMac Financial Services, Inc. faced a tumultuous financial landscape as 2023 was marked by significant challenges in the origination market. The industry volumes plummeted by roughly 40% from the previous year, hitting their lowest level since 1990. Despite this, the company demonstrated resilience, reporting a net loss of $37 million and a negative annualized return on equity of 4% for the fourth quarter. Contributing to these figures were nonrecurring accruals and net fair value declines on Mortgage Servicing Rights (MSRs) and hedges, amounting to a combined $234 million due to volatility in interest rates. However, removing the effects of these items reveals a robust underlying performance, with an annualized operating return on equity of 15%.
Against the sharp downturn in the mortgage industry, PennyMac managed to minimize the decline in its own production of mortgage loans to just 9%, showcasing its robust access to the purchasing market and its capability to support customers and business associates profitably under pressure.
The company's servicing segment was a pillar of strength, generating $268 million in pretax income, excluding the expenses related to litigation. Significant efforts in capital management also paid off as the company managed to raise over $1.5 billion in new long-term debt and redeemed $875 million in upcoming debt maturities. In a generous move, more than $110 million were returned to shareholders through dividends and share repurchases.
The quarter witnessed the final award of $150 million plus interest to Black Knight from an ongoing litigation. However, PennyMac retained proprietary rights to its SSE technology, a valuable asset that the company anticipates will open up more opportunities moving forward.
PennyMac has an optimistic outlook for the origination market, expecting that a trough was reached in 2023. This is based on declining mortgage rates and the anticipation of future rate reductions, which are estimated to elevate industry originations to around $2 trillion in 2024. This positions the company well, especially as its operating returns on equity have steadily climbed back to double digits over the year. Additionally, the servicing segment leads the earnings with an increase in operating pretax income due to the growth of PFSI's portfolio and higher short-term rates benefiting placement fees.
With a pretax income of $39 million in production for the fourth quarter, PennyMac anticipates building on this profitability in the forthcoming quarters. The expectation lies in the improvements to the origination market, potentially leading to enhanced financial results, despite some expected seasonal fluctuations in the first quarter.
The company succeeded in expanding its market share in correspondent lending to over 22%, up significantly from the previous year. This was paired with the continued growth of the broker direct channel, reflecting strong execution and technological edge. Additionally, the Board of Directors declared a fourth quarter cash dividend, reaffirming its commitment to shareholder returns despite the challenging market conditions.
The financials of the fourth quarter were largely influenced by the significant nonrecurring expense accrual related to Black Knight. Excluding this accrual, the Servicing segment's contribution to pretax income was $63 million, which underscores the segment's underlying profitability apart from extraordinary items.
You don't have any saved screeners yet
Good afternoon, and welcome to PennyMac Financial Services, Inc.'s Fourth Quarter and Full Year 2023 Earnings Call. Additional earnings materials, including presentation slides that will be referred to in this call are available on PennyMac Financial's website at pfsi.pennymac.com. Before we begin, let me remind you that this call may contain forward-looking statements that are subject to certain risks identified on Slide 2 of the earnings presentation that could cause the company's actual results to differ materially as well as non-GAAP measures that have been reconciled to their GAAP equivalent in the earnings materials.
I'd now like to introduce David Spector, PennyMac Financial's Chairman and Chief Executive Officer; and Dan Perotti, PennyMac's Financial's Chief Financial Officer. Gentlemen, I'll turn the call over to you.
Thank you, operator. Good afternoon, and thank you to everyone for participating in our fourth quarter earnings call.
PFSI reported a net loss of $37 million and an annualized return on equity of negative 4% in the fourth quarter. These results included a nonrecurring accrual of $158 million relating to our long-standing arbitration with Black Knight and $76 million of net fair value declines on MSRs and hedges given significant interest rate volatility during the quarter. Excluding the impact of these items, performance was very strong with an annualized operating return on equity of 15%, marking the culmination of another outstanding year for the company and highlighting the strength of our balanced business model.
2023 was one of the more challenging origination markets in recent history, with industry volumes down approximately 40% from 2022 and unit originations at their lowest level since 1990. However, PennyMac through its multichannel production platform generated $69 million of production pretax income and produced nearly $100 billion in UPB of mortgage loans, down only 9% from 2022. This demonstrates both our strong access to the purchase market and our ability to profitably support our customers and business partners.
These production volumes continue to drive the organic growth of our servicing portfolio which ended the year with more than 2.4 million customers and over $600 billion in UPB, up 10% from the end of last year.
Our servicing business generated $268 million in pretax income, excluding the Black Knight accrual. As the second largest producer of mortgage loans in the country and the fifth largest servicer, we have achieved significant scale in our mortgage banking platform with the capacity for continued growth and profitability in the years to come. This management team's ability to effectively manage capital has always been a competitive advantage for PFSI, and I am extraordinarily proud of the work we accomplished in 2023. Not only did we return more than $110 million to stockholders through share repurchases and dividends, we took meaningful steps to further strengthen the balance sheet, issuing more than $1.5 billion in new long-term debt at attractive terms and redeeming $875 million in debt with upcoming maturities.
I would like to provide a brief update on our long-standing litigation with Black Knight as outlined on Slide 5 of our earnings presentation. In January, the arbitrator issued a final award of $150 million plus interest to Black Knight, down from the interim award determined in November. PFSI recorded the related expense accrual in the fourth quarter, as previously noted. While we disagree with the ruling, we are very pleased with the arbitrator's affirmation that SSE remains our own proprietary technology as well as providing PennyMac the ability to utilize it as we see fit to benefit our customers and stakeholders.
Since we launched the system in 2019, it has performed extremely well, meaningfully enhancing our capabilities while helping to drive down costs. With this technology now free and clear of any restrictions on use or development, we believe there is potential for additional opportunities for the company and stakeholders over time.
Turning now to the origination market. We believe the overall market troughed in 2023 as mortgage rates have declined from their recent highs, and anticipated future rate cuts have increased third-party estimates for industry originations in 2024 to approximately $2 trillion. Much of this anticipated growth is based on expectations for interest rate reductions later on in the year, and we expect volume in the loan origination market to remain seasonally low in the first quarter of 2024 before moving into the spring and summer home-buying season. With a balanced business model and scale in both production and servicing, we remain very well positioned if interest rates remain high or decline further.
As you can see on Slide 7 of the earnings presentation, operating returns on equity have increased throughout 2023, returning to the double digits and consistent with our goals at the beginning of the year. The Servicing segment continues to drive earnings and operating pretax income has improved in recent quarters due to the growth in the size of PFSI's own portfolio and increased earnings from placement fees on custodial balances due to higher short-term rates.
Operating expenses remain low given our growing operational scale and continued low delinquencies. In production, while the market is expected to remain competitive, margins have improved over the course of 2023, and we estimate we have gained a considerable amount of market share, especially in the correspondent and broker direct channels, which provides strong access to the purchase market. As we add these higher note rate mortgages to our portfolio, we are creating additional opportunities for our consumer direct business to offer our customers a new lower rate mortgage when interest rates do decline. At year-end, 22% of our servicing portfolio consisted of mortgages with note rates in excess of 5%.
In the fourth quarter, production pretax income was $39 million. And while we expect some seasonality in the first quarter, we expect to build on this profitability in future quarters as the origination market improves. As I said earlier, I am extraordinarily proud of what we accomplished in 2023, and I am even more excited about PennyMac Financial's future. Our long track record of strong operational and financial performance is unique in the mortgage industry and has been driven by the resilience of our balanced business model with industry-leading positions in both production and servicing as well as our strong capital and risk management disciplines.
I believe we are the best-positioned company in the industry with a fully scaled balanced business model, proprietary industry-leading technology, a strong balance sheet and a growing number of servicing customers that stand to benefit from the products and services we offer to fulfill their homeownership needs.
I will now turn it over to Dan, who will review the drivers of PFSI's fourth quarter financial performance.
Thank you, David.
PFSI reported a net loss of $37 million in the fourth quarter or negative $0.74 in earnings per share for an annualized ROE of negative 4%. As David mentioned, these results include a nonrecurring expense accrual of $158 million before income taxes, or $2.20 per diluted share after income taxes related to the final award of our long-standing arbitration with Black Knight. PFSI's Board of Directors also declared a fourth quarter cash dividend of $0.20 per share. Book value per share was $70.52, down from the end of the prior quarter, primarily due to the net loss.
Turning to our production segment. Pretax income was $39 million, up from $25 million in the prior quarter. Total acquisition and origination volume were $26.7 billion in unpaid principal balance, up 6% from the prior quarter despite a decrease of more than 20% in the size of the origination market from the prior quarter. $24.2 billion was for PFSI's own account, and $2.5 billion was fee-based fulfillment activity for PMT. PennyMac maintained its dominant position in correspondent lending, with total acquisitions of $23.6 billion in the fourth quarter and margins similar to levels reported last quarter. We estimate that in 2023, PennyMac represented more than 22% market share in correspondent lending, up from 15% in 2022. We attribute this market share growth not only to the retreat of certain market participants, but also to our consistency in execution and industry-leading technology. Acquisitions in January are expected to total approximately $6.6 billion, and locks are expected to total $6.9 billion.
In broker direct, we see strong trends and continued growth in market share as we position PennyMac as a strong alternative to the channel leaders. Both locks and fundings for the quarter were down in the single-digit percentages from last quarter, less than the overall market, and margins were down due to higher levels of fallout as mortgage interest rates declined. The number of approved brokers at year-end was over 3,800, up 42% from the end of the prior year, and we estimate that we represented approximately 3.6% of originations in the channel in 2023. In January, broker direct originations were $600 million and locks were $1 billion.
In consumer direct, volumes remain low. But as David talked about, we remain well positioned given the number of customers we have added to the portfolio with higher mortgage rates. Production expenses net of loan origination expense were 8% lower than the prior quarter, primarily due to lower compensation accruals related to financial performance.
Turning to Servicing. The Servicing segment recorded a pretax loss of $96 million, primarily driven by the nonrecurring expense accrual mentioned earlier. Excluding the accrual, Servicing contributed $63 million to pretax income, down from $101 million in the prior quarter, primarily due to higher net MSR valuation-related declines. Excluding valuation-related changes in nonrecurring items, Servicing had very strong results with pretax contribution of $144 million or 9.6 basis points of average servicing portfolio UPB, up from $120 million or 8.6 basis points in the prior quarter. Loan Servicing fees were up from the prior quarter, primarily due to growth in PFSI's owned portfolio as PFSI has been acquiring a larger portion of the conventional correspondent production in recent periods.
Operating expenses declined also due to lower compensation accruals related to PFSI's financial performance. As expected, earnings on custodial balances and deposits and other income decreased $10 million from the prior quarter as balances declined due to seasonal property tax payments. Realization of MSR cash flows decreased $14 million from the prior quarter due to higher average interest rates for the majority of the quarter. EBO income remained relatively unchanged, and we continue to expect its contribution will remain low for the next few quarters, while interest expense increased from the prior quarter due to higher average balances of debt outstanding.
The fair value of PFSI's MSR decreased by $371 million during the quarter, driven by a decline in mortgage rates, which drove expectations for increased prepayment activity in the future. Hedging gains were $295 million, offsetting 80% of the decline in the MSR fair values. The net impact of MSR and hedge fair value changes on PFSI's pretax income was negative $76 million, and the impact on earnings per share was negative $1.05. The Investment Management segment contributed $1.9 million to pretax income during the quarter, and assets under management were unchanged from the end of the prior quarter.
Finally, on capital. Last quarter, we noted the October issuance of a 5-year $125 million term loan secured by Ginnie Mae MSRs and servicing advances. In December, we successfully raised $750 million in 6-year unsecured senior notes and subsequently retired $875 million of secured term notes due in 2025.
We'll now open it up for questions. Operator?
[Operator Instructions] We'll take our first question today from the line of Kevin Barker with Piper Sandler.
I just wanted to follow up on your comments about SSE and with the technology being free and clear from any restrictions. You also made the quote I believe there's a potential for additional opportunities and benefits for the company. Could you maybe expound upon what you can do with this technology, whether it's -- utilize it within PennyMac or maybe expand upon the use of that technology within PennyMac or outside of PennyMac?
Sure, Kevin.
First of all, SSE has been a great system for us as a servicer firm. We adopted it at the end of 2019. We were tested very, not tested, but we put it to use with COVID in 2020, and it performed very, very well in terms of being able to meet the needs of our borrowers and offering forbearances and modifications. And it's a system that we believe gives us a tremendous amount of competitive advantage. It's cloud-based, and it's in it with full integration from front end, back end and middleware components. Customers are able to access our advanced web, mobile and IVR solutions easily with high satisfaction rates. It's got unique workflow attached to it with state-of-the-art technology.
And look, I think it gives us a competitive advantage in the marketplace. And so in and of itself, it's something that I think we continue to see our servicing expenses get driven down since we adopted it, and it's something that has been meaningful to us in our evolution as a top 5 servicer.
As it pertains to unfortunate litigation, look, we're very happy for the lawsuit to be in the rearview mirror. And while I disagree with the final ruling, suffice it to say, the ruling clearly had more positive than negative. We were happy to retain ownership of our servicing system. We own SSE free and clear to use and as we see fit.
In terms of other opportunities, it's been 60 days since the ruling came out and we've been working to get this -- to really get this behind us. I've been very encouraged by opportunities that presented themselves and people approaching us, and we're beginning the exploration and evaluation process. And we don't feel any hurry to do something. We want to do something that's best for all stakeholders, but suffice it to say, it's something that is very unique in the industry and gives us a real advantage.
And then you produced a 15% operating return on equity this quarter. It seems like you have quite a bit of momentum going into 2024, not only within the origination channel, but also on the servicing side. Do you feel that the 15% ROE is the run rate now, and there is potential for significant upside in '24, particularly if we were to see a bigger origination market?
Look, I think that, as you pointed out, the operating ROE was 15% in the fourth quarter, which was up 13% from the third quarter. I'm expecting us to continue to build on the core Q4 results. There's going to be some seasonality in the first quarter, but the servicing has just been unbelievable for us. Profitability continue to remain strong in this high interest rate environment. The consumer is continuing to perform and stay current. We have continuing higher servicing fees and the growth in our own portfolio.
Look, on the production side, we were very profitable in '23. As I said, we were more profitable in '23 than we were in '22 on the production side. And given the trends I'm seeing in margins first and foremost, but then also trends that we're seeing in correspond and broker direct, I'm really encouraged that if we do see rate declines, especially in the back half of the year as the markets are projecting, and we can see our production business grows profitability.
Just to add on to that a little bit. To echo what David said, we saw those ROEs -- operating ROEs increase through 2023 move upwards consistent with how -- what we had hoped for, put forth at the beginning of the year where we really did see Q1 of last year as a sort of trough. It's not necessarily going to be in a totally straight line. There is some seasonality to this business if we're looking at Q1, although we do think there is upside -- upside potential from that 15% ROE in 2024 overall, especially as you noted, if and as people expect the market becomes larger -- the mortgage market.
Our next question comes from the line of Michael Kaye with Wells Fargo.
At the 2021 Investor Day, you said a 20% ROE in a more of a normalized environment? Just wondering what's been some of the pluses and minuses since that analysis back then. I know you bought out -- back a lot of stock during the pandemic period. It was like 20% of your share count. Correspondent and broker direct are likely ahead of plan, servicing is doing great too. Consumer direct probably has pushed out the opportunity a little bit. What I'm trying to figure out is 20%, is that still the right goalpost considering you just did 15% in a very challenging mortgage market?
I think the -- that still is the goalpost that we have. Obviously, we're generally striving to do 20% plus. That's part of David's mantra, and that's what we have our focus on. But we do think -- we certainly think that 20% is achievable in a more normalized market.
If we're looking at 2024 in specific, it's probably still under what we would consider a totally normalized market. To your point, if we're looking at sort of the positives, what's changed since that projection. The -- we probably outperformed even what we expected in correspondent, especially given the market size and -- but in terms of some of the growth of the servicing portfolio over that period of time -- over the total period of time is probably a little bit lighter than what we had expected just because the market has been smaller.
And then on the growth of the direct lending channels has just been a bit more constrained because of the overall size of the market and especially the refi market in terms of consumer direct. So we still think 20% is a reasonable goal and an achievable goal. We still have our balanced business model where we have servicing as sort of the core and production on top of that, which helps drive up those overall ROEs, but that's -- we think that the overall balanced business models, what led us to that 15% ROE here in the fourth quarter, we think we can continue to build on that, but 20% is still sort of our target.
Okay. I wanted to talk about the competitive landscape in correspondent, looks like things slowed down a little bit in January. And I've been hearing some other large nonbanks like Freedom talking about getting more competitive after raising a large amount of corporate debt. I was just looking for an update on what's happening in correspondent so far this year?
Look, I think that -- look, we have a dominant position in correspondent. We have very deep relationships. And I would start off by saying that the market in January is going to be smaller than even in Dec. And I think that we're seeing good activity in correspondent, specifically with builders. A lot of our share growth that in the latter part of the year is because of the activity coming out of builders where we have very deep relationships, and that's something that really helped drove a lot of the share growth.
As you know, banks have been stepping back and we're just about at the point where I'm going to not being able to say that because they're going to be out. Sellers are not retaining servicing. And one of the things that we're also witnessing is there's a flight to quality, and we're seeing more and more sellers delivering a disproportionate amount of their loans to [ us ].
Interestingly enough in correspondent, we saw a decrease in the number of sellers. It's down to 812 from 830. And so we're seeing a little bit of consolidation taking place. But we -- I think that our value proposition is pretty tried and true. We're in the market every day. We get value for that. I think when others are in and then they're out and they're in and then they're out, that creates some consternation with sellers. And so this is why I continue -- we saw some really good margin growth in correspondent. We were at 21 basis points in the fourth quarter of '22. We were at 34 basis points in the fourth quarter of '23. And it's a combination of increasing pricing power, but also we're able to find executions away from the GSEs that provided some additional margin.
But I think that -- I like that the correspondent is so core to our franchise and it's something that I continue to expect it to be. But yes, I mean people raise more capital to get more aggressive. And on the margin, you may have someone sell loans to them because they have a little better bid. But we're the industry leader for a reason and will continue to be there.
Our next question comes from the line of Eric Hagen with BTIG.
I hope you guys are well. Yes, I thought the hedging results were good, but what do you feel like maybe prevented you from hedging even more of the fair value mark during the quarter? Is there a sense maybe for how much sensitivity you feel like there is to, call it, the next 50 basis points lower in mortgage rates with respect to the fair value mark?
Sure. So we're pretty pleased with our hedging performance in the fourth quarter. As you know, interest rates were all over the map in the fourth quarter. We were up -- the 10-year was up as high or a little bit over 5% and down below 4%. So really a pretty significant swing in terms of the direction, the direction that mortgage -- that interest rates and mortgage rates went through the quarter. We were able to cover 80% of the move and that included the costs that we have to hedge, which ate up a little bit as well. We have been in this environment, targeting a tighter hedge ratio than we have historically, but overall, given, like you said, the volatility of interest rates, we're pretty pleased with the performance. To your point, as we move -- as interest rates or if interest rates move down further, we have -- the asset has increasing sensitivity. Part of that is also because we've been adding a fair amount of loans at those higher interest rates that's been part of our strategy so that those loans that we have a greater portfolio of loans that could move into refinanceable territory if and when interest rates do decline. And so the sensitivity of the servicing asset has increased a bit as we've moved lower in interest rates and we continue to hedge that.
But what we've also seen consistent with our strategy is an uptick in refinances here in the first quarter. You can see that in terms of some of our January locks in the -- in our consumer direct channel, for example. We're up pretty meaningfully from the run rate that we had in Q4 and we've just taken another sort of leg lower here in interest rates. So really, again, gets back to the balanced business model and how, to the extent that we see potential slightly faster prepayment speeds on the servicing side with our hedge and with the refinanceability of the [ port ] that serves to offset that.
Yes. No, that's helpful. So how are you guys -- a separate question here, I mean, how are you guys thinking about the secured financing for MSRs versus maybe replacing some of that balance with unsecured debt? And do you see any risk that banks could pull back from supporting the market for MSR funding? And does that in any way kind of drive your appetite for unsecured?
So we've -- I mean, we've talked about it in the past that we really -- our strategy generally is to move more toward unsecured financing. Our issuance in Q4 $750 million of unsecured debt we used to pay off some of our secured financing or MSR financing, although really are really term notes there as opposed to bilateral from banks. Overall, we expect to continue to move more towards the unsecured financing. Really the reasons for that other than continuing to sort of bolster our credit position and move sort of a more favorable position with the rating agencies, where that unsecured debt is sort of more stable isn't subject to margin call -- margin calls if we have a significant interest rate rally, for example, so has some benefits on that side. We think we can drive down the costs over time as we move more towards unsecured since that has a more favorable sort of ratings and capital profile stability.
In terms of the -- flipping to the other side on the secured, we have not seen a pullback in MSR financing from banks. In fact, it's really been the opposite. We've seen more banks really willing to lend on MSRs and in different MSR structures. So -- and we've been, over time, adding banks to our MSR facilities, and so we don't see that as an issue. We want to continue to diversify the number of banks that we have that are financing our MSRs just for risk management purposes. But we don't see a pullback there and that that's not really a major driver of our move toward unsecured debt. It's really all of the other motivations that I discussed previously.
Our next question is from the line of Bose George with KBW.
Can you give us just your updated thought on share buybacks just given the current valuation?
Sure. So in terms of share repurchases, we didn't have any share repurchases in the fourth quarter. The share price has moved up pretty significantly from a few quarters ago as I'm sure you're aware. And so any time we're looking at our capital deployment, we're looking at what the relative return and relative value is in deploying it in shares versus really back into our business and continuing to acquire MSRs through correspondents and add to the servicing portfolio. And so we really see that as the sort of the optimal path currently.
And the other piece that plays into this is our overall management of the leverage ratio of the company. And so continuing to manage our leverage ratio and that a bit above 1 point or 1x in terms of our nonfunding debt to equity is where we've been. We're looking to managing that in a similar range there. And so obviously, share repurchases puts a little bit of pressure on that.
So those are the factors that we're balancing. But in the current environment, certainly in the fourth quarter, and given how everything is currently situated, we'd expect to continue with our capital deployment really into -- back into the business in MSRs as opposed to share repurchases.
Okay. Great. Makes sense. And then actually just switching over a follow-up on the MSR hedging question from earlier. You noted that you're going to run a tighter hedge ratio, I think you said 100% last quarter on the call. If volatility abates, I mean, should we have a sort of a more matched sort of hedge result, assuming we don't see sort of a big pickup in prepayments?
So in the first quarter, so I said we are seeking to manage to a bit of a tighter hedge ratio. What we've seen so far in the first quarter, again, a fair amount of volatility and an inverted yield curve, both of which are negative in terms of our hedge costs. And as we continue to get further down into lower territory with a higher percentage of our portfolio and higher mortgage rates, won't necessarily be hedging quite as close to 100%. In the fourth quarter, we were at 80%. So I think you could expect us to be -- we won't necessarily be right at 100% in terms of our hedge ratio,. still higher than we've been historically, but in terms of the cost and where the portfolio is positioned, a little bit of potential for some differences between the MSR and hedge in the first quarter.
Our next question comes from the line of Mark DeVries with Deutsche Bank.
First, a question for Dan on the margin. I heard you comment on the quarter-over-quarter decline in margin in broker direct being attributable to the fallout effect. Didn't hear whether you commented on the decline in consumer direct. Was it the same fallout effect? Or is there something else that pushed again sale lower in consumer direct?
Sure. In consumer direct, really what was influencing the margin there is a shift toward more refinances as opposed to second-lien origination. So second-lien originations have a smaller balance. And so on a per unit basis, although the dollars of revenue per loan for a refinance is higher, the basis points are lower because the loan size is larger. So it really -- that's really what's driving the margin decline in basis points from Q3 to Q4, is given the interest rate rally that we saw toward the end of Q4, we started seeing a pickup repo and reacted. And we've seen that continue here in the first quarter with a further decline in interest rates. But it's really -- on a unit basis, actually, we're seeing the revenue per unit go up. But on a basis points basis, we're seeing those declines.
Okay. Got it. And then a follow-up for David on the new opportunities around your servicing platform. How do you think about the trade-offs between potentially sharing a source of competitive advantage with competitors versus the incremental revenue that you might be able to generate off of that?
Look, I think -- but from our perspective, these are all the questions we're asking ourselves as we come out of the litigation. I think that there is demand in the marketplace for more competition. It's not good or healthy for the industry to have reliance on any one person. And given the market share of the leader out there, there are many in the industry who share my point of view. I think from our perspective, we have to answer that question from an economic value, are we best served by continuing to maintain the competitive advantage that we're seeing in our servicing platform and in terms of driving down costs and increasing productivity versus what we can get by other means. But as I said earlier, this is very early on in the whole process, and this is something that we're going to -- we're working already on it, and this is something that we're just going to continue to work on. And as we always do, we're going to get to the right place.
Our next question comes from the line of Kyle Joseph with Jefferies.
Just want to pick your brain a little bit more on the correspondent channel. So the number of sellers obviously went down, but what do you see that doing to margins over time? Obviously, that impacts supply and not necessarily demand, but I was just curious to get your thoughts on how that impacts your margins in that segment?
Yes. I don't think it's really any impact on margins. It's really sellers who weren't selling us a lot of loans, and we weren't active in the marketplace. I think margins are -- margins have been pretty stable over the past couple of quarters. And even in January with volumes down, we're seeing margins stable, albeit they're up very nicely from where they were a year ago.
Look, I think that, of course, we always like higher margins, and we're -- I always remind you people to get more margin. But I think that Doug and Abby and the team do just a phenomenal job continuing to support the operation in terms of the value proposition that they provide to our correspondent sellers. And look, we're on -- where loan prices are today, sellers can't really -- can't afford to keep servicing. And so a lot of the alternative execution over the years has been to the GSE's cash window where they would either retain the servicing for a lengthy period of time or ultimately sell it. And so I think we're getting more loans coming in through the whole loan channel.
And as I said, there's just a natural flight to quality when you see a marketplace where there's -- where you have people going in and out and you got people getting out of the business. And we've been at this every day since we got into it almost 15 years ago. And it's something that gives us a halo effect in correspondent and it allows us to maintain both margin and share.
Got it. Very helpful. And then on just expenses, the outlook for '24. Obviously, you guys have been doing a good job of getting more efficient in a tough market, but would you expect those to kind of stay in parallel with volumes? Or how much variability is there in that line item -- sorry, particularly on the production segment.
On the production segment, to your point, we did a lot of our work even going back to '22 in terms of bringing down the expense base. That served us really well in '23. As David noted on the production side, we actually had higher pretax income in the production segment in '23 than we did in '22, and that was really due to getting our expense base rightsized. . As we -- in most of the rest of the business, if we're talking about sort of the core functionality, the corporate and shared services as well as the servicing side. We expect those to be fairly stable as the servicing portfolio grows. There will be some additional expense but to the extent that we see the sort of soft landing and not significant increases in delinquencies, we expect those to grow really relatively slightly given some of the efficiencies that we have in that business and on the corporate side to be very contained.
And then on the production side, to the extent that we see an uptick in production, there will be some uptick in the production expenses that goes along with that, and that will really be determined by the size of the market. And in particular, on the interest rate decline on the refinance side and if that expands, that would necessitate some additional growth in those expenses.
Our next question is from the line of Trevor Cranston with JMP Securities.
One more question on the hedging of the servicing side of things. The earnings on the custodial balances have obviously become pretty significant over the last several quarters. Can you talk about any hedges you guys have put in place to sort of protect that earnings stream as that fund potentially starts to move lower? And also just talk in general about how we should sort of think about the impact of lower Fed funds on the economics of the servicing business.
Sure. So our -- the -- custodial balances, the earnings on the custodial balances on our servicing portfolio are projected to really follow the forward curve. So to the extent that rates today are projecting a decline in short-term interest rates, which they are, our projection for those cash flows that's embedded in our MSR value reflects lower earnings on those custodial balances as we move forward -- as we move forward in time. And that -- those changes since the forward curve would change if you shock interest rates up or down, those changes in the earnings on the custodial balances are also incorporated into our hedge. That's part of what we see changing in the value of servicing if we were to see an interest rate shock down. And that's part of what we're hedging for, hedging against. If you look at -- and we've sort of, I think, put that out there in our earnings materials that our custodial balances, our earnings on custodial balances are generally tied to short-term interest rates or Fed funds to the extent that we see Fed funds go down, we'd expect the rate that we're earning on those custodial balances to follow to some extent -- to follow relatively closely the same way that it followed as interest rates increased. We'll also see some of our financing costs decrease as the portion of our portfolio -- the portion of our financing that's really secured by MSRs is generally floating rate. And so you have a bit of an offset there as well in terms of the -- that revenue versus the expense. But on the hedging side -- so those will be somewhat offsetting. But on the hedging side, that's part of what we are hedging against for a decline in interest rates is the decline in that -- in those earnings on custodial balances.
Our next question is from the line of Shana Chao with Bank of America.
Previously, on the call, you mentioned the margin calls under secured debt. Like how should we think about how much rates need to decline before you see any impact of margin calls on your secured facilities?
We're pretty over-collateralized at the moment. So currently, at the end of Q4, we had a little under $1 billion of cash on the balance sheet. We have the ability to draw against our secured facilities for around $2 billion of additional value. And as I mentioned, we paid down some of our secured facilities during the quarter with our unsecured debt issuance. So we really have a pretty -- it would have to be a pretty significant reduction in value to get to a point where we would have a margin call on our unsecured debt. . Additionally, the hedges that we have in place, if we have a decline in interest rates, generate cash and that cash can then be used to pay off any margin calls that we might have. And so we're really both from an over-collateralization point of view as well as from sort of the performance of the hedges, we're pretty significantly covered off against the risk of the margin impact.
Okay. Great. That's helpful. And then I think we've heard from some third parties that even first season books, there's been delinquencies for certain pools of Ginnie Mae over 10%. It looks like the 60-day delinquency increased 40 bps sequentially for the USDA, but still relatively low at 5.2%. Can you just comment on what you're seeing and then generate delinquencies and expectations going forward?
Overall, we've seen our Ginnie Mae delinquency fairly stable. There's always some amount of seasonality as you're going through the fourth quarter. especially in the Ginnie Mae book will you have an uptick towards December. And usually, there's a pretty meaningful downtick in February and March as folks receive their income tax refunds. And so we saw a bit of that overall on the portfolio -- overall in the portfolio, though, delinquencies have been pretty contained as they were through the whole year. We publish our overall delinquency profile for the MSR in the deck were up slightly from the prior year in terms of delinquencies, overall delinquencies.
So it's not something that we see as a significant issue. I agree that in certain pockets, there has been some pressure, and we have seen some in certain areas, some delinquencies, uptick. Similarly, we've seen probably better-than-expected performance in others.
The other piece that I mentioned, which is the way that delinquencies impact us. You may have seen that our servicing advances increased quarter-over-quarter. Really, that was not driven very significantly by delinquencies. That was really driven by seasonal property tax payments. A lot of that increase was really from current borrowers where just property values have appreciated pretty significantly over the past few years. Property tax amounts change, that impacts the amount that's being escrowed and then maybe a shortage for a payment or 2 while the escrow analysis is redone and the borrower's escrow account sort of catches up.
If you actually look at our servicing advances year-over-year, they went down slightly year-over-year in terms of looking at Q4 to Q4. So the summary of all that is just that delinquencies, we have not seen a significant shift. And in terms of how it impacts us on a cash basis has been very contained and very similar to last year.
We have a final question today from the line of Kevin Barker with Piper Sandler.
I just wanted to follow up on the amortization -- the realization of cash flows line that came down pretty meaningfully. And we also saw prepay speeds drop quite a bit this quarter. Now obviously, seasonality played a big part in that. But do you feel that the realization of cash flows remain fairly low as we would go through the first half of 2024 given the portfolio is producing very low prepay speeds at this time?
To your point, the realization of cash flows declined a bit going from Q3 to Q4 as interest rates were high for most of the fourth quarter. We have seen a decent interest rate decline as we go into Q1, as I mentioned there, we have seen an uptick -- some uptick in refinances that will flow through to some uptick in prepayment speeds. And so I think we do expect some uptick in the realization of cash flows as we're going into the first quarter, given those dynamics. But overall, we expect that there's -- for the servicing portfolio and servicing profitability will still be significant and meaningful as we're going through the next year.
Great. And then could you just provide maybe a little bit more depth on the demand for refinances versus closed-end seconds. Now I realize the demand is very low relative to the overall market and what it has been in the past, but just given your customer base in the servicing portfolio, are you seeing your customers start to lean in more towards refinances in last month or two or are you seeing the closed-end second still starting to garner more attention?
Look, as rates decline, we do see more customer lean in on closed-end seconds, primarily on the servicing that we added in 2023. And that's -- excuse me, pre 2023 for closed-end seconds. On cash outs, I think, look, I think that as rates -- as rates decline, you'll see more borrowers lead into cash outs versus closed-end seconds. And as rates stay higher, they're going to be leaning into closed-end seconds. It's -- for us, it's the appropriate product mix that we have as it gives borrowers the opportunity to take cash out of their -- take equity out of their property typically to pay down lower cost debt. . But we're in a position right now where the rallies in the marketplace still have not as meaningful effect as the majority of the mortgage market is, as you know, in kind of 3% and 4% mortgages. So I think that you will see an increased amount of rates. You'll see an increased amount of refi as rates do decline. But look, the closed-end second product for us has been great for consumer direct in terms of maintaining capacity in place. We've launched it to our non-port customers, which allows us to do profitable closed-end seconds and maintain capacity in place for when we do see a significant decline in rates in consumer direct.
We've also introduced it in broker direct. And I think that that's something that I'm encouraged. And look, the brokers have the need to maintain capacity just like we do in consumer direct, and it's a product that best serves their customers. It also adds to the broker direct value proposition that we have. And I'd be remiss if I did bring up that look, we're continuing to gain share broker direct. And I think we're starting to see more and more brokers seeing PennyMac as a strong alternative to the top 2 participants. And in addition, we've invested a lot in technology to support the brokers. And so I think that we like what we're seeing in broker direct and margins are maintaining their levels. And so as closed-end second versus cash out refi versus rate and term refi, we've got to cover it in our production divisions, and we're going to participate no matter what the rate environment is.
And just to add on to what David said and to address one part of your question. So the refinances that we're seeing in that shift that I had talked about is really shifting our -- really our resources to addressing rate and term refinances primarily in -- at the end of Q4 and Q1 as opposed to the second lien because that is really a more profitable product for us and also something that isn't necessarily persistent depending on what happens to interest rates. And so to David's point, to the extent interest rates go up, we have the second lien product and expanding breadth of that to be able to move into to the extent interest rates decline, we have the ability more loans will be refinanceable. It wasn't -- it isn't a shift. I want to make sure that I didn't give the impression that we were shifting from second -- doing second lien to doing cash out refis. We didn't kind of cross over that threshold. It's really rate and term refinances are where we saw the uptick in Q4 and now in Q1.
We have no further questions at this time. I'll now turn it back over to Mr. Spector for closing remarks.
Thank you. I want to thank everyone for joining us this afternoon. We went over a lot of good information, and we had what I thought was a really outstanding quarter. I want to thank everyone for the thoughtful questions. If anyone has any additional questions, feel free to reach out to our Investor Relations team by e-mail or phone. And again, thank you all for joining the call.