Back in 2009. Over time, the mortgage market has evolved, and the overall focuses of both PMT and PFSI have evolved as well. As PFSI grew to a full-service mortgage company involved in both originating and servicing loans, as well as with significant expertise in the overall mortgage industry in terms of its management team, PMT also evolved. Its partnership with PFSI is really focused around generating organically created investments for PMT, where PMT benefits from PFSI's ability to service the loans, to originate and underwrite and oversee the acquisition of the loans, to package and securitize the loans, and to hold, and then PMT holds those securities. PMT, with its lower cost of capital, with its tax-advantaged status as a REIT, being able to hold those mortgage investments, really benefits from that relationship with PFSI.
In today's environment, PMT is really focused on three different strategies. The first strategy that PMT is focused on is what we call our credit-sensitive strategies. That is really around mortgage investments whose primary risk is credit risk. Typically, subordinate pieces of securitizations that hold mortgages. Historically, and the most significant part of our credit-sensitive strategies currently is what we call credit risk transfer. That is taking credit risk, the most subordinate portion of credit risk related to loans that have been securitized with Fannie Mae, with the government-sponsored entities. These are high credit quality loans, but exposure to the initial parts of the losses around those loans.
That program that we had with Fannie Mae is no longer in place, really ran through the end of 2020 in terms of generating new assets of that type, where specifically those loans all came from PennyMac's production, came through PennyMac's system, were sold and securitized with the GSEs, and then the credit risk is held by PMT. Those loans generally have significant seasoning and really high credit quality, low current loan-to-value ratios, and we feel those are a very solid investment. Those comprise about 15% of the total equity of PMT. In addition to that, and we'll get deeper into this currently, our current focus for PMT's credit-sensitive strategies is really around generating new securitizations of high credit quality loans.
Most recently, we've been doing a securitization a month of non-owner-occupied loans that are otherwise eligible to be delivered to the GSEs, so high credit quality from that perspective. Again, taking the most subordinate credit risk of those securitizations. We'll get further into that and some of the dynamics around that a little bit later. The second strategy that PMT focuses on is our interest rate-sensitive strategies. That's really seeking to create a strategy where we are earning the return from investments that are primarily sensitive to interest rates. However, we run a very sophisticated hedge program to insulate ourselves from changes in value of the assets, from a significant amount of changes in value of the assets that are sensitive to changes in interest rates.
The largest part of our interest rate-sensitive strategy is our mortgage servicing rights, and those comprise about 45% of our total equity. Again, those are assets that PennyMac was involved in creating. The loans came through PennyMac's production, were sold and securitized generally with the GSEs, again, high credit quality loans. The servicing is also the act of servicing those loans is also performed by PFSI, and so high amount of ability to oversee the credit performance of those loans. Those mortgage servicing rights behave in terms of their sensitivity to interest rates really inversely to the way that most other interest rate-sensitive assets behave.
As interest rates go up, the value of mortgage servicing rights increases because you expect the loans, the underlying loans, to pay off more slowly, which is the opposite of what you see from most other interest rate-sensitive assets where if interest rates increase, their value will decrease. To offset that, we have a portfolio of high credit quality mortgage-backed securities and other interest rate hedges that we'll go into a little bit later. The purpose of the offsetting assets in our hedging program is really to significantly insulate ourselves and our book value from changes in interest rates through time and to maintain a more stable earnings profile or overall earnings profile in terms of those interest rate-sensitive assets. The third strategy that PMT is engaged in is correspondent aggregation.
This is really the aggregation of loans from smaller originators that PMT holds on its balance sheet and then sells and securitizes primarily with the GSEs historically, although, as I mentioned, in more recent periods have been aggregating those for securitizations that are not with the GSEs where we're holding the subordinate credit risk. That strategy is in partnership with PFSI. PFSI also is engaged in the correspondent lending program, and PMT, given that relationship, is able to sort of dial up or dial down the total amount of correspondent activity that it wants to perform based on its capital availability and its objectives with the underlying loans. That really covers the scope of PFSI.
With those three different strategies—credit-sensitive strategies, interest rate-sensitive strategies, and correspondent—all of those, as I mentioned, really rely and are benefited from the relationship with PFSI, who has the expertise to be able to both select the assets as well as to generate the assets generally through the correspondent activity and service the underlying loans of the majority of the assets.
Okay. Great. Thanks for that overview. In terms of the—can you talk about just the current size of the equity and asset base? You mentioned the asset mix. How do you see that evolving?
Sure. The overall equity base is about $1.9 billion. About $560 million of that is preferred equity. In terms of common equity, about $1.3 billion-$1.4 billion of common equity. If you look at the overall balance sheet in terms of the assets, we have about $15 billion in assets as of 3/31. However, we do have, because of the securitizations that we're performing, a number of—or all of those securitizations are consolidated on balance sheet, despite the fact that a lot of the debt related with the securitizations is really non-recourse. All of the senior debt, apart from the subordinate pieces that we keep that we've sold out into the market, is non-recourse, is only supported by the assets that are in those securitizations or the loans that are in those securitizations.
If you look at the assets that are not directly supporting that non-recourse debt, it really takes that down to—or the total non-recourse debt is about $3 billion. If you look at the offsetting assets, that comes out to about $12 billion of total assets.
Okay. Great. In terms of sort of places you want to increase your allocation of capital, can you discuss that? You've obviously talked a lot about the securitization opportunity in non-owner-occupied. Can you just walk through that asset class, the opportunity, the risks, etc.?
Yeah. Great. So, yeah, as Dan mentioned, within the credit-sensitive strategies, we've been most recently focusing on organically creating investments through the retention of subordinate tranche securitizations in the non-agency securitization market. Essentially, the process is through the mortgage bank at PFSI and the ownership of its share of the correspondent business within PMT, loans that are originated that are either agency eligible or are strictly non-agency. PennyMac either sells those whole loans directly to investors or to banks. In the case of what we've been doing since November of last year, we've been aggregating specific types of collateral and issuing them as non-agency deals on a monthly basis. More specifically, it's been with agency-eligible investor and second home collateral. The deal size has been roughly about between $300 million and $400 million a month.
We sell off all of the senior tranche risk to end investors, and we retain approximately about, call it $25 million-$30 million of the first loss pieces, the subordinate tranches, and then lever those and kind of get a, call it low to mid-teens type return. This is certainly, given the synergies that Dan talked about between the two companies, a very repeatable process. We service the collateral ourselves. It is definitely a good story in terms of the connectivity between the portfolio managers within PMT that are managing the assets that are on the balance sheet and connectivity with our servicing area to ensure best outcomes if there are any type of credit events. Just in terms of attractiveness of the asset, when we stress it from a default standpoint and severities as well as prepayments, very stable asset.
We are looking to expand into other collateral types. Jumbo borrowers, for example, is kind of the next area that we are looking. Like I said, given the synergies, it is a very scalable investment. We have been really excited about it. The reception from the investor community has been strong.
Okay. Great. In terms of the credit profile of those assets, can you talk about that, especially versus CRT where the performance obviously has been very good over the years?
Yeah. So I mean, with this borrower type, tends to skew certainly better in terms of credit profile, for example, than say if you were to look at some non-QM securitizations, a little bit more lower credit profile. So we think the credit profile is very clean. That certainly, when you look at loss severities within kind of investor loans over time, that bears out. Certainly within the jumbo space, these tend to be quite high credit quality borrowers. Generally in the jumbo space, that is really prepayment risk that you are levered to, right? Because for a given incentive, those really large loan sizes are going to prepay quite fast. From the standpoint of owning the first loss piece, if what I care about is credit more than prepayment speeds, then even subordinate tranches on prime jumbo collateral is quite attractive.
Compared to, you mentioned that the performance within the CRT investments has been strong. That's true. Dan mentioned this earlier. In terms of where we sit today, the accumulated HPA is significant just because we're talking about 2017 to early 2020 vintage borrowers that are sitting on a lot of equity in their homes. They've delivered considerably. They have current LTVs probably in like the 50-ish type area.
Okay. Great. Actually, just one more on this topic. Can you just talk about the competitive landscape for this product?
Yeah. Sure. Right now, we at PennyMac are, I think, taking out a pretty significant share of investor and second home collateral within the correspondent origination business just overall. That is a function of the team and kind of the focus on building out market share there. We have, I think, probably one other competitor in terms of issuing non-agency investor/second home back deals. They issue quarterly around $300 million, and we will call it monthly for a bit more than that. The prime jumbo space is much more competitive, just a lot more players in that area, just a lot more origination as well as you would expect. I would say there is within each, both prime jumbo as well as investor second home, respectively, pretty homogeneous type deals. It is really a function of investor selecting for servicer type for the most part.
There tends to be, because there is more issuance in prime jumbo space as well as what I mentioned earlier in terms of the prepayment risk associated for the senior most tranches, the spread volatility tends to be a bit higher in the prime jumbo space than, say, investor second home.
Okay. Great. Just switching to the MSR strategy, can you just talk about that opportunity also in the context of the Rocket deal? Is there any—do you see more need for growth or just—yeah, talk about the impact of trends in that market and what you guys are seeing.
Yeah. You go ahead.
Sure. Overall, in terms of the MSR strategy, as I had mentioned, MSR is around 45% of PMT's total portfolio. A significant portion or the vast majority of those assets were really originated in 2020, 2021 at lower rates. They have less sensitivity to prepayments than loans that are originated more recently. We do like for PMT really that balance and that concentration because there is more stability of the cash flows. They are more predictable. They are generally less sensitive to volatility and interest rates, which we have seen a significant amount recently. Really, we like the overall composition of PMT's portfolio near the levels that it is and so are not necessarily totally focused on growing that portfolio very significantly for PMT currently. If you look at the overall MSR portfolio over the last few quarters, it has been around the same level to slightly declining.
That is really our general focus for PMT going forward, is most likely maintaining a level of stability there and really more focused, as Marshall mentioned, on further investments on the credit side. If you look at our overall interest rate-sensitive strategies versus credit-sensitive strategies, closer to or a little under 60% of our equity is invested in our interest rate-sensitive strategies versus a little under 30% in our credit-sensitive strategies. As we continue with the securitizations and that activity, we think that that is our attractive assets to be adding with stable profiles over time. We think that that is really more—or that is really more of the focus for PMT currently.
How do the returns and the rate strategy now compare to the, I guess, a low to mid-teens return on the credit side?
The interest rate-sensitive strategy overall, the returns have been a bit more compressed in recent periods, really as a function of the overall inverted. When I say inverted, I'm really talking about short-term rates to longer-term rates to flat profile of the yield curve. The interest rate-sensitive strategies, since typically we are invested in assets that have longer lives, whose overall unlevered returns are based on longer parts of the yield curve, and then financing them with either short-term or floating-rate debt, typically those strategies will perform better when the yield curve is a bit steeper, when longer-term asset yields are a bit higher and shorter-term financing rates are a bit lower. Given the inversion that we'd seen in recent periods and now relative flatness puts a bit more compression on our expectation for the interest rate-sensitive strategies as a whole.
The overall return on equity that we've seen there or expect there in that kind of environment is a bit lower, low double digits generally. We do see upside in that strategy to the extent that we see a steepening in the yield curve, either with longer-term rates moving higher or shorter-term rates moving lower. Since we are hedged in terms of the assets of that strategy, we shouldn't see a significant market value impact or book value impact if we do see that kind of a move or if we see longer-term rates moving higher. It would improve our overall go forward earnings profile or we expect in that kind of a situation.
Okay. Great. Thanks. I just wanted to switch over to risk management. Historically, you guys have done a good job of keeping your book value steady, especially relative to many of the peers. Can you talk about your strategy for managing your book value and just your approach to that?
Yeah. Sure. I think it's a function of a couple of things. The asset mix that we've been talking about, so not just being wholly concentrated in one investment area. Running mid-single digit type leverage obviously helps as well. As Dan was talking about for our MSR investments, ensuring that we have stability from a hedge coverage ratio standpoint, meaning that for very large moves in interest rates on a standalone basis, the MSR asset can change in value quite a bit. Trying to mitigate that through hedges has been certainly key in ensuring that we have book value stability. The daily kind of decision-making process really centers around how much you want to hedge out. You have an asset that exhibits what we would refer to as negative convexity. In order to mitigate that, you're buying derivatives.
You're buying options that you're paying premiums for. It's trying to ensure that the cost of hedging is not so overly punitive or substantial that it weighs down on the returns.
Okay. Great. Just from the risk management standpoint, I wanted to switch over to the financing side of it. Can you just talk about the lack of sort of essentially margin calls in most of your funding, sort of the benefits of that?
That has been a really key other prong in our risk management strategy and something that really worked well for us in times of stress, most recently during the pandemic. Really, the two prongs, the interest rate hedging, as we talked about, as well as ensuring the stability of our funding. With our credit risk transfer program, which is sensitive to changes in credit perception, one of the key focuses that we had going into the pandemic as we were accumulating that asset is to ensure that we had debt that would not be impacted by mark-to-market changes. The majority of our credit risk transfer position is financed by term notes that do not have mark-to-market, that do not have mark-to-market positions or do not have mark-to-market provisions, rather.
In addition to that, we also finance part of the balance sheet with unsecured debt, both exchangeable notes as well as baby bonds that we've issued into the market. That covers a lot of the risk on the credit-sensitive side where we don't have hedges against changes in credit perception that could impact the value of the assets. That was, as I mentioned, a really great benefit during the pandemic where we saw a really sharp drop in assets that were credit-sensitive during the onset of the pandemic. Several other entities that held similar types of assets were forced to sell their assets into a very weak market and sort of crystallize those losses. Whereas given our debt position and the non-mark-to-market aspect of it, we saw the recovery of that and the recovery of our book value since we're not forced sellers in that type of a market.
The other prong that Marshall mentioned is our interest rate hedging. When interest rates are volatile or as interest rates are volatile, given that we have these interest rate-sensitive either offsetting assets or positions, although most of our debt is sensitive to margin call risk there because we have the offsetting assets where we can either call cash or we have lines where we can pull down cash if our asset values increase or our hedge positions are immediately realizable into cash. When interest rates move significantly, we have similar protection both from a book value side as well as from a liquidity perspective.
Those are really, I would say, the two prongs of the risk management side that have helped keep us in good stead and ensure that if we do see impacts to book value, generally speaking, they're not crystallized given the underlying good fundamental credit quality of the underlying assets.
Okay. Great. Let me just ask a quick one on the dividend, and then I'll open it up if there are questions. Can you just talk about the comfort level with the dividend?
We have maintained our dividend at around $0.40 per share for a really significant number of quarters now going back to the pandemic. Overall, we see great value in dividend stability. Our run rate that we publish on a quarterly basis has been slightly below the dividend. As I mentioned, given the potential on the interest rate-sensitive side for the curve to steepen and increase the returns in our interest rate-sensitive strategies and the continued investment in the credit-sensitive strategies and the shift there, we feel comfortable maintaining the dividend at the $0.40 level.
Great. We have a little under a minute left. Any questions from the—go ahead, Eric.
How do you guys run the market today with some debt capital? Can you talk about what you're using that for and where it takes your leverage and how you think about leverage going forward?
We can't comment on that, but thank you for the question.
Anything else? Actually, we've got 20—I was going to ask on GSE privatization, but probably not. Ten seconds is probably not enough. Thanks very much, and thanks very much to the audience for joining us.