july 23, 2020 mtgage sategy my - fhn financial

FHNFINANCIAL.COM800.456.5460

Disclaimer is on the last page of this report.

MORTGAGE STRATEGY MONTHLY

FHNFINANCIAL.COM800.456.5460

MORTGAGE STRATEGIESWalt Schmidt, CFA

[email protected]

Alexis Vilimas312.258.5066

[email protected]

Brandon Messing312.258.5057

[email protected]

JULY 23, 2020

Targeted Approach to Payups p. 2The unprecedented actions by the Fed to affect borrower-friendly policies in the 1-4 family mortgage market are causing some unique challenges for MBS investors. Last month, we highlighted the extreme yield impact these policies are having on large portions of the market. In this missive, we elaborate a bit more with some guideposts investors can use to better identify value. – Walt Schmidt

Appraising Appraisal Waivers p. 10 The share of new loans receiving appraisal waivers has grown dramatically over the past few months, reaching a record high of 40% in June. In this section, we review recent changes to appraisal polices and discuss the implications of appraisal waivers on prepayments for TBA collateral and specified pools. – Alexis Vilimas

Market Update p. 22 Month-to-date, conventional 30yr and 15yr MBS production coupons generally outperformed UST and swap hedge ratios. CMO spreads are 15-25 bps tighter since June. Hybrid spreads and CMO Floater DMs are 2bps tighter. Payups for loan balance specified pools are unchanged since the end of last month. The discount for Jumbo conforming paper decreased by 2-8 ticks for 2.0s-3.5s. Conventional fixed prepayment speeds increased across the coupon stack in June, with seasoned lower coupons increasing by more than 20%. The increases were due to lower primary rates, strong seasonal turnover rates, and two additional business days.

July 23, 2020 of 25

MORTGAGE STRATEGY MONTHLY | TARGETED APPROACH TO PAYUPS

Executive Summary: The unprecedented actions by the Fed to affect borrower-friendly policies in the 1-4 family mortgage market are causing some unique challenges for MBS investors. Last month, we highlighted the extreme yield impact these policies are having on large portions of the market. In this missive, we elaborate a bit more with some guideposts investors can use to better identify value.

The Fed has pushed MBS asset prices to almost-record highs while maintaining a cost-of-funds policy very close to the zero-lower-bound (ZLB, or 0%). So not only are prices near historical highs, but more importantly, the ratio of coupon to yield in the MBS market, a more direct measure of premium risk, is at historic highs. That creates an environment in which ever more vigilance is required at the cusip level to ensure as much carry as possible for the portfolio – including the mandate to avoid negative carry. This was the subject of the first article of the June, 2020 edition of the Mortgage Strategy Monthly.

This month’s journey within the realm of the MBS market takes us deep into the specified pool market to find more specific guideposts to look for in order to maintain carry. There are three basic means that investors can utilize to maintain carry in the MBS book in the face of higher dollar prices and faster prepayment speeds:

1. Down-in-coupon trades.

2. Extension of amortization schedules (i.e. sell 15yr, buy 30yr).

3. Pay up for collateral characteristics to improve call protection (i.e. LLB, 100% NY, LTV, etc.)

There is, of course, a fourth and more risky means that some MBS investors can utilize to improve carry in the current environment: the MBS roll market. Rolls in all of the production coupons (generally 2.0s through 3.0s) in the TBA market are trading special, which allows investors to carry those coupons at an assumed financing rate that is below market. This is due in most part to the aggressive bid by the Fed for the front-month production coupons relative to the back month. Some investor mandates are very used to this transaction, but it is not without risk. If the roll ceases to become special or if the investor needs to exit the position for other reasons, the investor will likely be delivered a very disadvantaged pool relative to the dollar price, as we alluded to in last month’s missive.

Due to this risk, most institutional investors seek to earn alpha or carry relative to their liabilities utilizing one or more of the three methods described above. For a number of years, we have published a rich/cheap model for the more popular “pure play” payup stories across the coupon stack. The summary output for that is displayed in Table 1 below.

https://docs.fhnfinancial.com/?5211ae9e-0190-4e0d-afba-d2c2a96c2af7

July 23, 2020 of 25


Table 1: Payups Very High, but Now “Cheap”

Source: FHN Financial

These payup stories are the most popular and after a few months of very low rates and very high dollar prices for TBA, payups are starting to come off a bit. That is not to say that payups are low, by any means. In fact, they are still quite high on an absolute basis. But relative to the historic low of the 30yr Freddie Mac commitment rate, payups have increased at a less aggressive pace lately. This can be seen by using the LLB 3.5 payup as a proxy in Figures 1 and 2 below.

Pure play payups flashing “cheap”, but absolute levels still high.

July 23, 2020 of 25


Figure 1: Current Fit Plot for LLB 3.5 Payup vs. 30yr Freddie Survey Commitment Rate


Figure 2: Time Series of LLB 3.5 Payup and Rich/Cheap Model Valuation


Using LLB 3.5 as a proxy, then, payups are “cheap” based on the model but by no means “low”. That begs the question that, even if one could find an LLB 3.5 (difficult to do), is the greater-than 6-point payup for LLB 3.5 something one would entertain? That would put the dollar price at almost $112.

View of the LLB 3.5 payup in the context of primary mortgage rates.

Payups still historically high, but “cheaper” relative to rates.

July 23, 2020 of 25


Perhaps a more reasonable hypothetical would be something like a HLB 3.5 at a payup of around 4.5 points or approximately a $110 dollar price. With a Gross WAC approaching 4.50%, a pool like that would be in-the-money by about 150 basis points. But even that would be a hypothetical, because payup stories very seldom trade as “pure plays” in reality. Much of that has to do with servicer profiles. Many non-bank servicers are now involved in the creation and sale of payup story bonds, because they command a premium. However, they typically command less of a premium than do bonds serviced by large regional banks, which tend to prepay slower (think Truist). Therefore, we will consider a generic HLB 3.5 and use an S-curve to determine different prepayment assumptions.

The data in Figure 3 below are an S-curve that plots prepayments for various loan balance limits relative to refi incentive. In order to make sense of the data, the chart also plots a “universe” cohort that is all of the loans within the 6-36 WALA 30yr conventional cohort, as well as a “cheapest-to-deliver” proxy for TBA. The definition of that proxy is displayed on the chart.

Figure 3: Compare Prepayments of CTD to Max Loan Balance Cohorts

0

5

10

15

20

25

30

35

40

45

50

55

-150 -125 -100 -75 -50 -25 0 25 50 75 100 125 150 175 200

1 M

onth

CPR

Refinance Incentive

S-Curve Analysis by Loan Balance | Loan Level DataLast 12 Months | 30yr Conventional MBS, 6-36 WALA

CTDUniverse201-225k176-200k151-175k111-150k HLB86-110k MLB0-85k LLB

Cheapest to Deliver: Nonbank Servicer, LnSz>225k, LTV<75, and FICO>740

Source: FHN Financial and CPRCDR

The lines on the chart represent where the historical one-month CPR has been over the past 12 factor reports for those max loan balance categories by incentive bucket, along with the universe and CTD cohorts for comparison purposes. In order to determine whether the 4.5-point payup for HLB (i.e. 150k max) collateral is worthwhile for a 3.5 coupon, we will compare yield and carry profiles at the spot where that coupon is “in-the-money”, approximately 150 basis points. (The GWAC on most 3.5s is around 4.50%, and the current

Loan balance stories produce the most efficient call protection in the MBS market.

July 23, 2020 of 25


Freddie 30yr survey rate is right at 3.00%.) To use round numbers, we will assume 55 CPR for the CTD 3.5 and 25 CPR for the HLB 3.5 bond. Also, to assume round numbers, we will use $105-16 for the CTD and $110, the buck, for the HLB 3.5. The yield and spread advantage of the HLB 3.5 is displayed in Figure 4 below.

Figure 4: Very Negative Yield and Spread Profile for CTD 3.5 at Current Speed

Source: FHN Financial, CPRCDR, Bloomberg

Figure 5: Much Better – But Still Not Great – Yield and Spread for HLB 3.5 Proxy


To be sure, we are simply using the generic 3.5 cash flow as a proxy. The CPR inputs in each case correspond to the actual historical data for the two cohorts, CTD and HLB. The CTD proxy can prepay as slow as around 5 CPR and as fast as 55 CPR. In fact, that is where it is currently paying, so that is the scenario we highlight in Figure 4. By contrast, the HLB cohort pays as fast as 25C, where it is currently paying, so we highlight that scenario in Figure 5. Although the yield and spread in the base case are not particularly attractive for the HLB 3.5, at least they are positive, unlike the CTD category!

A couple of other things to note about these two profiles:

1. There is more cushion in the CTD profile in the backup scenario, because even at a premium, the dollar price is so much lower than for the HLB bond.

2. There is room for the prepayment speed of the HLB 3.5 to accelerate by 10 CPR to 35 for it to display a “breakeven” yield and spread profile relative to the CTD 3.5.

Higher coupons with no call protection currently produce negative yields.

150k max helps, but absolute dollar price is extremely high.

July 23, 2020 of 25


Although interesting, we find ourselves a bit dissatisfied with the analysis. After all, why do all of this work for a bond that yields only 37 basis points? Furthermore, the HLB 3.5 bond carries less on a monthly basis than does the CTD 3.5 even at the slower 25 CPR prepayment scenario. Is it possible to find a story that is a lower payup but that still offers important call protection? The answer is, “yes”, but it is imperative to peel back more layers of the prepayment onion.

As we mentioned above, there are three ways to add duration/call protection to the portfolio. The third method is highlighted above. But what if we employ both methods 1 and 3? That is, what if we both go down-in-coupon and add call protection collateral? In addition, we can add less than a “pure play” convexity story to make the payup – and hence the dollar price – more manageable.

To make the comparison relevant to the foregoing, we will stay with the 150k max Loan Balance story. But there will be two offsetting features to consider. First, the trade will be down-in-coupon into the 30yr 2.5 sector. Second, the substantial call protection offered in 150k will be offset by Quicken as servicer. This will simultaneously reduce the dollar price and produce less call protection. The important question is: how much less, and is the lower price enough to compensate?

Figure 6: A Twist on 150k Loan Balance Prepay Performance

0

5

10

15

20

25

30

35

40

45

50

55

-150 -125 -100 -75 -50 -25 0 25 50 75 100 125 150 175 200

1 M

onth

CPR

Refinance Incentive

S-Curve Analysis by Loan Balance | Loan Level DataLast 12 Months | 30yr Conventional MBS, 6-36 WALA

CTD

Universe

150k Max, Quicken

150k Max, Complement

Cheapest to Deliver: Nonbank Servicer, LnSz>225k, LTV<75, and FICO>740


Offsetting collateral characteristics produce less than a “pure play” call protection story.

July 23, 2020 of 25


The data in Figure 6 are similar to those in Figure 3. Each graph displays the prepayment response by refi incentive for the “universe” and the “cheapest-to-deliver”, or TBA, cohorts. The difference is that Figure 6 breaks out the 150k max loan size cohort between those loans originated by Quicken (in blue) and the complement (in red). The Quicken cohort offers a good deal of call protection relative to that of the CTD, or TBA, cohort, but less than that of the complement 150k cohort.

Applying these actual prepay results to a yield table format produces the outcome described in Table 2 below. This is not a TRR/OAS view of the world. This simply looks at yield and spread outcomes based on a range of expected prepayment outcomes, based on each security’s particular collateral characteristics. It is also not a traditional +/-300 basis points view of the world. After all, down 300 is very unlikely since we are at the zero lower bound in the short rate. Also, prepayments do not change much once the market moves beyond 150 basis points in either direction. In fact, burnout takes quicker effect and prepays actually begin to slow for the +175 and +200 bps incentive buckets, as displayed in Figures 3 and 6 above.

Table 2: The “Cheap” Form of Call Protection Provides Value Across Prepay OutcomesPool Number Description Coupon GWAC WAM Price

FN CA6592 150k Max 2.50 3.11 359 107.00 CPR: 5 8 10 12 18 24 30Yield: 1.74 1.55 1.43 1.29 0.85 0.38 -0.14

Avg Life: 10.68 8.38 7.25 6.34 4.51 3.41 2.70I-Spread: 111 104 97 90 61 18 -31

FR QB1858 150k Max Quicken 2.50 3.43 353 105.44 CPR: 5 10 15 20 25 30 40Yield: 1.90 1.65 1.37 1.07 0.75 0.41 -0.35

Avg Life: 10.67 7.25 5.30 4.09 3.28 2.71 1.95I-Spread: 127 120 107 85 57 24 -50

FN MA4096 Major Pool 2.50 3.37 358 104.69 CPR: 5 10 20 30 40 50 60Yield: 1.98 1.76 1.26 0.68 0.01 -0.75 -1.66

Avg Life: 9.05 6.47 3.88 2.65 1.95 1.50 1.17I-Spread: 135 131 103 51 -14 -90 -180

Expected Prepay Distribution (CPR)


All three of these bonds have between 25 and 50 basis points of refinancing incentive, so we drew a box around that outcome from the S-curves in Figure 6 to represent the “base case”. The other scenarios represent the rest of the S-curve distribution for each particular collateral type. It is certainly much more satisfying to view the comparison this way rather than the antiquated “street consensus” that treats all three of these collateral types equally.

The 150k max Quicken pool is displayed in the middle of the analysis, because it is a middle-of-the-road approach to earnings and risk management. It has a yield profile in the base case that is close to that of the higher-payup pool and it is much better than the major pool in the base case. The 150k max Quicken pool beats the fully-priced 150k pool in the slower CPR environments, and it easily beats the major pool in the higher CPR environments. It even does well vs. the full-payup pool in the faster CPR scenarios.

Targeted approach on lower coupons with offsetting characteristics is best option.

July 23, 2020 of 25


Some key strategy takeaways from this analysis:

1. Payups for “pure play” payup stories are still very high historically, even though they have come off a bit lately and are now flashing “cheap” on our regression model. This is a function of the overall level of rates (i.e. funding costs near 0%) and the Fed’s involvement in the TBA markets (i.e. taking some of the worst-to-deliver pools off the market’s hands).

2. We find value in less than “pure play” call protection stories. The higher the dollar price, the greater the cost of being “wrong” on the prepay assumption. Also, there is a natural ceiling to both MBS dollar prices and payup levels in a market with amortizing assets. Some of the “pure plays” are trading well over $110. Pools with offsetting features such as loan balance issued by “faster” servicers are cheaper ways to onboard call protection with quantifiable risks.

3. For investors who are bearish on rates, do not buy any call protection. The current payups will only eat into returns in a back-up. In fact, rate-bearish investors should not even consider 30yr 2.5s. They should move into 15yrs and up-in-coupon on the 30yr stack with no call protection. Given where dollar prices are, there is a good deal of cushion available in those sectors.

July 23, 2020 of 25

MORTGAGE STRATEGY MONTHLY | APPRAISING APPRAISAL WAIVERS

Executive Summary: The share of new loans receiving appraisal waivers has grown dramatically over the past few months, reaching a record high of 40% in June. In this section, we review recent changes to appraisal polices and discuss the implications of appraisal waivers on prepayments for TBA collateral and specified pools.

In March, Fannie Mae and Freddie Mac released a new loan disclosure called Property Valuation Method that identifies loans with and without appraisal waivers. Following the release, we published our initial review of the disclosure in the March FHN Mortgage Strategy Monthly. We concluded that loans with appraisal waivers pay 6-10 CPR faster than loans with traditional appraisals even after controlling for differences in incentive and collateral characteristics. We also predicted waiver usage to increase gradually as more lenders adopt the practice to help alleviate capacity constraints.

Since then, refinancing has surged and prepayments have consistently printed faster than expected, showing essentially no COVID-19 related slowdown. (In fact, Covid-19 likely contributed indirectly to faster prepayments as a result of the Fed’s aggressive purchases of MBS.) Appraisal waiver share has grown dramatically in line with the surge in refinancing activity. Figure 1 shows the share of loans receiving an appraisal waiver since June 2017. Last month, 40% of all new loans received an appraisal waiver.

Figure 1- Time Series of Appraisal Waiver Usage

0%

10%

20%

30%

40%

50%

60%

0%

10%

20%

30%

40%

50%

60%

Jun-17 Dec-17 Jun-18 Dec-18 Jun-19 Dec-19 Jun-20

% P

rope

rty In

spec

tion

Wai

ver (

PIW

)

Loans Originated with an Inspection Waiver by Purpose | Single Family UMBS

Non-Cashout Refi Cashout Refi Purchase Total Issuance

Loan Purpose Jan-18 Jan-19 Jan-20 Jun-20Non-Cashout Refi 12% 14% 46% 59%

Cashout Refi 0% 0% 11% 24%

Purchase 2% 4% 6% 8%

Total 7% 6% 21% 40%


When borrowers have significant financial incentive to refinance whether or not their appraisal is waived, as they do now, does the PIW flag still inform prepayments? As more loans receive waivers, is it still a useful indicator of faster future speeds? The following analysis is a continuation of our initial review, highlighting the impact of appraisal waivers on recent prepayment speeds.

Appraisal waiver share has increased significantly in 2020.

https://my.fhnfinancial.com/ref/111890/Pages/p/mortgagestrategy.aspx?ItemId=24860

https://my.fhnfinancial.com/ref/111890/Pages/p/mortgagestrategy.aspx?ItemId=24860

July 23, 2020 of 25


At the end of March, the GSEs announced two important changes to appraisal policies. First, the GSEs announced temporary flexibility of their “full” appraisal standards to adhere to social distancing guidelines. The new guidance allows exterior-only appraisals for non-cashout refinances for GSE owned loans. The new guidance also states that exterior-only appraisals or desktop appraisals are acceptable for purchase loans below the 85 LTV threshold. Cashout refinances still require a full traditional appraisal, including an interior and exterior report. Although this change does not affect appraisal waivers directly, it does add to a climate that encourages lenders to adjust standards for refinancing.

The second announcement directly affects the decision process for appraisal waivers. Freddie Mac relaxed their loan eligibility requirements for appraisal waivers. Cashout refinances were excluded previously from the program under Freddie. Now, appraisal waivers can be applied to cashout refinances for loans with up to a 70 LTV for primary residences and up to a 60 LTV for second homes. Freddie also increased the maximum LTV for non-cashout loans for primary residences from 80 LTV to 90 LTV. Below is a summary of the eligibility criteria for loans to receive a waiver in both programs.

Figure 2- GSE Program ComparisonCriteria for Appraisal Waiver Programs

Fannie Mae Freddie MacProgram Property Inspection Waiver Automated Appraisal ValuationsSystem Desktop Underwriter (DU) Automated Collateral Evaluation (ACE)Start Date January 2017 June 2017Property TypeSingle Family 1 unit properties Eligible EligibleCondominiums Eligible Eligible2-4 unit properites Ineligible IneligibleProperties with values > $1 million Ineligible IneligibleCo-ops units and manufactured homes Ineligible IneligibleProperties with resale restrictions Ineligible IneligibleMortgage TypePurchase

Primary residence 80 LTV 80 LTVSecondary residence 80 LTV 80 LTV

Investment property 80 LTV IneligibleCash-out Refinances

Primary residence 70 LTV 70 LTV (Previously Ineligible)Secondary residence 60 LTV 60 LTV (Previously Ineligible)

Investment property 60 LTV IneligibleNon Cash-out

Primary residence 90 LTV 90 LTV (Previously 80 LTV)Secondary residence 90 LTV 80 LTV

Investment property 75 LTV Ineligible

Source: Fannie Mae, Freddie Mac and FHN Financial

In June, Freddie’s appraisal waiver share increased 10% from 33% to 43%, surpassing Fannie for the first time. Freddie's appraisal waiver share may be increasing more than Fannie's appraisal waiver share due to their recently relaxed LTV eligibility criteria.

Fannie and Freddie use similar criteria to determine waiver eligibility.

July 23, 2020 of 25


Figure 3- Time Series of Appraisal Waiver Usage by Agency

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

Jun-17 Dec-17 Jun-18 Dec-18 Jun-19 Dec-19 Jun-20

% P

rope

rty In

spec

tion

Wai

ver (

PIW

)

Property Inspection Waiver Share by Agency | Single Family UMBS Origination

Fannie Mae Freddie Mac

Agency Jan-18 Jan-19 Jan-20 Jun-20Fannie Mae 8% 6% 25% 38%

Freddie Mac 3% 5% 18% 43%


Which eligible loans are most likely to receive a waiver? It is not surprising that loans that receive waivers have many of the same characteristics that generally lead to faster prepayments. Higher loan size loans have higher waiver shares than lower loan sizes. Broker loans have a 44% waiver share, more than 10% higher than the other origination channels. States in the western half of the country tend to have a higher percentage of waiver loans. More than 40% of loans in California, Colorado, Washington, Arizona, and Oregon received appraisal waivers since July 2019. New York, Florida, and Texas have the lowest share of waiver loans.

Non-banks have much higher waiver shares than large banks and increased their waiver shares by more than large banks in the past four months. The ten servicers with waiver shares over 30% are all non-banks. Most notably, over 50% of all Quicken and United Shore issuance since July 2019 received an appraisal waiver. Below is a summary of appraisal share by collateral characteristic.

Freddie waiver share surpassed Fannie for the first time last month.

July 23, 2020 of 25


Figure 4- Share of Appraisal Waiver Loans by Characteristic

July 2019 - June 2020 Issuance<=80 LTV, excludes Investor loans

20 Largest Servicers by Current Balance 20 Largest States by Current BalanceLoan Purpose %Bal %Waiver Servicer %Bal %Waiver State %Bal %WaiverPurchase 26% 12% Quicken 9% 50% CA 21% 43%Cashout Refi 27% 14% Wells 9% 29% TX 6% 20%Non-Cashout Refi 47% 57% Chase 6% 18% FL 5% 19%

United Shore 5% 53% WA 5% 41%Loan Balance %Bal %Waiver PennyMac 5% 33% CO 4% 44%85k 1% 10% New Residential 4% 37% AZ 4% 42%110k 2% 17% Lakeview 3% 30% IL 3% 39%125k 2% 20% Calibur 2% 38% NY 3% 14%150k 4% 23% Truist 2% 17% MA 3% 36%175k 5% 26% US Bank 2% 20% NJ 3% 31%200k 6% 28% Matrix 2% 31% VA 3% 38%225k 6% 32% Amerihome 2% 25% MI 3% 39%300k 21% 35% Loan Depot 2% 45% NC 3% 22%400k 24% 39% Pingora 2% 25% GA 3% 32%>400k 28% 36% Citizen 1% 27% UT 2% 37%

Home Point 1% 35% OR 2% 40%Origination Channel %Bal %Waiver Mr. Cooper 1% 40% MN 2% 33%Broker 14% 44% Specialized Loan 1% 35% PA 2% 28%Correspondent 30% 29% PNC 1% 31% OH 2% 27%Retail 56% 33% Fifth Third 1% 26% MD 2% 36%


Which MBS products have the largest share of waivers? The share of waiver loans in 15yr and 20yr pools is elevated compared to the share in 30yr pools. In general, there is a higher appraisal waiver share in refi loans and a larger percentage of refi loans in 15yr and 20yr pools. Fannie pools have a slightly larger share of waivers in all products due to a slightly longer history of their program. Fannie rolled out its appraisal waiver program six months ahead of Freddie Mac and had less restrictive LTV requirements, until recently.

Figure 5- Waiver Usage in Securitized Products

Outstanding % Waiver % Outstanding % Waiver % Outstanding % Waiver % Outstanding % Waiver %FN 30yr 46.3% 4.1% 51.8% 6.3% 49.2% 16.0% 48.2% 28.8%

FH 30yr 33.2% 0.9% 34.0% 3.0% 37.7% 10.7% 33.4% 27.1%

FN 15yr 8.4% 9.7% 5.5% 12.5% 5.1% 26.9% 7.8% 40.8%

FH 15yr 6.1% 1.7% 4.2% 5.4% 3.7% 18.4% 5.4% 39.7%

FN 20yr 3.4% 9.5% 2.7% 11.8% 1.5% 32.8% 2.9% 44.0%

FH 20yr 1.8% 2.5% 1.2% 6.2% 1.2% 19.5% 1.8% 42.6%

FN 10yr 0.7% 9.4% 0.4% 11.2% 1.5% 32.8% 0.4% 36.5%

FH 10yr 0.1% 1.6% 0.1% 6.2% 0.1% 17.8% 0.1% 37.2%

2017 Vintage 2018 Vintage 2019 Vintage 2020 Vintage


Across coupons, lower coupons have significantly higher percentages of appraisal waivers due to the higher composition of low friction refi loans. Figures 6 and 7 show the percentage of appraisal waivers in FN major pools since 2017. Last month, 42% of the balance in 30yr 2.5 major pools and 19% of the balance in 30yr 3.0 major pools received appraisal waivers. In 15yr major pools, 58% of 2.0s and 42% of 2.5s were originated without traditional appraisals. Also worth noting is that although the waiver percentage has trended down for 30yr 3.0-3.5s and 15yr 3.0s since the beginning of the year, the overall percentage is still considerably higher than pre-2019 origination.

Higher waiver share in larger loans, broker origination, non-bank lenders, and western states.

Waiver share is higher in 15yr than 30yr.

July 23, 2020 of 25


Figure 6- Appraisal Waiver Time Series in 30yr Majors

0

5

10

15

20

25

30

35

40

45

50

0

5

10

15

20

25

30

35

40

45

50

Jan-17 May-17 Sep-17 Jan-18 May-18 Sep-18 Jan-19 May-19 Sep-19 Jan-20 May-20

% W

aiver

% Property Inspection Waiver in 30yr FN Major Pools

2.5 3.0 3.5 4.0 4.5 5.0Source: FHN Financial and CPRCDR

In our Mortgage Strategy 2020 Outlook, we suggested that 2019 collateral was potentially more negatively convex than 2018 collateral due to specific collateral characteristics. The 2019 vintage had higher WAC spreads, larger loan sizes, higher LTVs, higher refi share, lower investor share and higher broker share than previous years. These are all features that imply faster prepayment speeds. During a period of historically low rates, the combination of these characteristics and a higher share of appraisal waiver loans heightens prepayment concerns for lower coupon pools.

Lower coupons have significantly higher share of appraisals.

July 23, 2020 of 25


Figure 7- Appraisal Waiver Time Series in 15yr Majors

0

10

20

30

40

50

60

70

0

10

20

30

40

50

60

70

Jan-17 May-17 Sep-17 Jan-18 May-18 Sep-18 Jan-19 May-19 Sep-19 Jan-20 May-20

% W

aiver

% Property Inspection Waiver in 15yr FN Major Pools

2.0 2.5 3.0 3.5 4.0Source: FHN Financial and CPRCDR

How much faster do loans with appraisal waivers pay compared to loans with traditional appraisals? Figures 8 and 9 show prepayment speeds for 2018 and 2019 vintage 30yr 3.5s since January 2019 by loan purpose. The graphs restrict the loans to low friction borrowers to isolate the effect of appraisal waivers on speeds in non-spec pools. On average, purchase loans with a waiver prepay 6-10 CPR faster than purchase loans with an appraisal. In the last four months, the spread between waiver and appraisal loans increased slightly, with waiver loans prepaying 8-13 CPR faster.

58% of 15yr 2.0s and 42% of 15yr 2.5s received waivers.

July 23, 2020 of 25


Figure 8- Purchase Loan Prepayment History

0

10

20

30

40

50

60

70

80

0

10

20

30

40

50

60

70

80

1 M

onth

CPR

Appraisal vs. Appraisal Waiver Speeds | 30yr Fixed ConventionalPurchase Loans | Low Friction Borrowers

2018 3.5, Appraisal Waiver

2018 3.5, Appraisal


2019 3.5, Appraisal

Low Friction- <=80 LTV, >225k, >720 FICO


Speeds accelerated in the 2H 2019 and again in 2Q 2020 for non-cashout refinancing overall. Starting in July 2019, 2018 3.5s waiver loans printed speeds 7-10 CPR faster than appraisal loans. Non-cashout refi loans with waivers in 2019 3.5s printed much faster speeds than appraisal loans, between 15-20 CPR faster for most of the time series. However, recently the differences have compressed which may suggest that the impact of PIW lessens some into a refi wave. Last month, 2019 3.5s waiver loans printed speeds 9 CPR faster and 2018 3.5s printed speeds 7 CPR faster than appraisal loans.

Purchase loans with waiver prepay 6-10 CPR faster.

July 23, 2020 of 25


Figure 9- Non-Cashout Refi Prepayment History

0

10

20

30

40

50

60

70

80

0

10

20

30

40

50

60

70

80

1 M

onth

CPR

Appraisal vs. Appraisal Waiver Speeds | 30yr Fixed ConventionalNon-Cashout Refi Loans | Low Friction Borrowers


2018 3.5, Appraisal


2019 3.5, Appraisal



S-curves provide a more complete picture of the prepayment landscape for appraisal waiver loans. Appraisal waiver, purchase loans print about 5 CPR faster for loans with 50-75bps of incentive and 10 CPR for loans with 100bps of incentive or more. Appraisal waiver, non-cashout loans print faster speeds across all incentive buckets, ranging from 5 CPR faster for discount coupons to 20 CPR faster for premium coupons.

Speed differences have decreased slightly over the past four months.

July 23, 2020 of 25


Figure 10- Speed Comparison by Incentive

0

10

20

30

40

50

60

70

80

90

100

0

10

20

30

40

50

60

70

80

90

100

-50 -25 0 25 50 75 100 125 150 175

1 M

onth

CPR

Refinance Incentive (bps)

S-Curve Analysis | July 2019- Current6-18 WALA | 30yr Low Friction Borrowers

Purchase, Appraisal Waiver

Purchase, Appraisal

Non-Cashout Refi, Appraisal Waiver

Non-Cashout Refi, Appraisal



Last month in the June FHN Mortgage Strategy Monthly, we proposed that originators are shifting their focus to segments of the market that have not participated in the first stage of the refi wave. The increase in the share of appraisal waivers in spec loans is additional evidence that the market is moving into phase two of the refi wave. The share of waivers in spec loans in 2Q 2020 is close to the same share of waivers in non-spec from 1Q 2020. NY, Investor, and LLB loans have the lowest share of waiver loans. Lower payup spec loans have the highest percentages of appraisal waivers, specifically higher loan balance stories and low FICO paper.

Non-spec appraisal waiver loans prepay 5-10 CPR faster than appraisal loans.

July 23, 2020 of 25


Figure 11- Appraisal Waiver Usage in Story Collateral

0%

5%

10%

15%

20%

25%

30%

35%

40%

Non-Spec 0-85k 86-110k 111-150k 151-175k 176-200k 201-225k Low FICO NY FL TX Investor

PIW

% o

f New

Issu

ance

Waiver Share by Specified Story | 30yr UMBS Origination

2019 Q3 2019 Q4 2020 Q1 2020 Q2Source: FHN Financial and CPRCDR

Figure 12 shows the speed differences between waiver and appraisal loans for specific collateral features. Generally, the same collateral stories that have the highest waiver percentage also have the highest speed differentials. Waiver loans that are also low FICO, 225k max, and 200k max print speeds 8-14 CPR faster than appraisal loans with the same characteristics and incentive.

Lower payup story loans have the highest shares of waivers.

July 23, 2020 of 25


Figure 12- Speed Comparison by Characteristic, Waiver Loans vs. Appraisal Loans

0

2

4

6

8

10

12

14

16

0 50 100 150

1 Mon

th CP

R Di

ffere

nce


Speed Differentials by Story | Waiver Loan Speeds Less Appraisal Loan Speeds30yr UMBS, 6-18 WALA | Last 3 Months

0-85k

86-110k

111-150k

151-175k

176-200k

201-225k

Low FICO

NY

FL

TX

Investor


Slicing the data by servicer exposes another pattern. Non-bank serviced loans appear to have larger speed differentials than bank serviced loans for story collateral. Figure 13 compares the speed differentials between the two largest servicers, Quicken and Wells, for the three largest collateral features by waiver percentage, low FICO, 200k max, and 225k max. At almost every level of positive incentive, Quicken has larger speed differentials than Wells. Although more analysis is needed, it is reasonable to suggest that waiver share is more meaningful for non-banks than large banks in specified pools.

Low FICO, 225k max, and 200k max print speeds 8-14 CPR faster with waivers.

July 23, 2020 of 25


Figure 13- Speed Comparison by Characteristic and Servicer, Waiver Loans vs. Appraisal Loans

0

5

10

15

20

25

30

35

40

0 25 50 75 100 125 150 175

1 Mon

th CP

R Di

ffere

nce


Speed Differentials by Servicer | Waiver Loan Speeds Less Appraisal30yr UMBS, 6-18 WALA | Last 3 Months

Low FICO Quicken

Low FICO Wells

200k Max Quicken

200k Max Wells

225k Max Quicken

225k Max Wells


In summary, even in a time of exceptionally low rates and high media effect, the PIW flag informs future prepayments. Below are the key points from this analysis:

�� With positive incentive, non-spec loans with appraisal waivers pay 4-10 CPR faster than non-spec loans with traditional appraisals.

�� Low FICO, 225k max, and 200k max loans with appraisal waivers print speeds 8-14 CPR faster than traditional appraisal loans with the same characteristics and incentive.

�� Non-bank serviced, lower payup story loans, have larger speed differentials between waiver loans and appraisal loans than their bank serviced counterparts.

�� We expect waiver usage to continue to increase as more lenders adopt the practice to help alleviate capacity constraints.

Quicken has larger speed differentials between waiver and appraisal story loans.

July 23, 2020 of 25

MORTGAGE STRATEGY MONTHLY | MARKET UPDATE

Z-Score* Week MTD YTD High Low Avg

Prices

30 Year 3.0 105.30 1.4 (0.30) 0.06 6.38 105.84 100.66 102.863.5 105.42 1.3 0.31 0.25 3.98 106.11 101.50 103.704.0 106.30 1.2 0.41 0.27 3.42 106.94 103.19 104.814.5 107.59 1.2 0.28 0.05 3.58 108.44 104.56 106.205.0 109.53 1.7 0.30 0.19 4.23 109.63 106.02 107.705.5 110.44 1.7 0.00 0.17 3.52 110.53 107.16 108.68

15 Year 2.5 104.84 1.6 (0.23) 0.20 3.91 105.16 100.38 102.283.0 104.97 1.2 (0.14) (0.19) 2.45 105.64 101.66 103.403.5 105.09 1.0 0.14 (0.02) 1.28 105.73 103.16 104.294.0 106.19 1.8 0.42 0.38 1.69 107.00 103.88 104.854.5 105.08 1.2 0.08 1.39 2.27 106.50 101.89 103.755.0 104.86 0.5 0.20 0.81 (0.91) 106.56 100.53 104.125.5 109.45 1.0 0.19 0.22 2.61 112.00 98.25 105.54

20 Year 2.5 104.67 1.5 (0.20) 0.27 2.53 104.88 99.09 101.813.0 105.69 1.5 (0.19) 0.23 1.77 106.13 101.70 103.533.5 105.41 0.9 0.02 0.08 0.61 106.42 103.28 104.524.0 106.31 0.9 0.14 0.13 0.20 107.03 104.17 105.454.5 107.52 0.8 0.20 (0.19) 0.59 108.69 105.09 106.725.0 109.31 1.5 0.08 (0.03) 1.50 109.63 106.02 107.64

I-Spreads (UST)

30 Year 3.0 53.9 -1.9 (1.1) (19.5) -20.8 158.0 43.3 85.33.5 77.4 -1.2 (4.6) (17.9) -4.6 195.4 70.0 97.04.0 70.5 -1.7 (10.8) (23.5) -18.8 200.5 70.2 98.04.5 68.1 -1.2 (5.6) (14.0) -16.0 212.1 60.8 89.35.0 73.0 -0.6 (11.9) (13.5) 3.3 191.4 53.2 85.45.5 126.0 0.5 (7.5) (17.9) 38.3 174.5 51.0 106.86.0 98.0 -0.2 (6.6) (20.2) 7.5 143.9 53.4 104.36.5 178.7 0.2 (6.7) (9.2) 46.4 226.1 96.0 170.9

15 Year 2.5 16.0 -1.9 4.7 (17.1) -24.8 151.9 8.3 45.73.0 48.3 -0.2 4.3 (0.6) 13.5 178.6 32.8 52.03.5 75.8 1.1 (3.1) (9.6) 36.9 175.3 27.4 56.34.0 44.5 -0.4 (4.8) (9.1) 8.3 154.1 25.5 52.7

Mortgage Rates

Conforming 30 Year 3.15 -2.8 0.01 (0.15) (0.71) 4.12 3.13 3.6615 Year 2.68 -2.6 (0.05) (0.11) (0.73) 3.44 2.67 3.105x1 Hybrid 3.19 -1.3 0.11 0.06 (0.30) 4.25 2.87 3.62

Borrower Activity

MBA Refinance Index 3,973 1.0 199 614 2,114 6,419 1,375 2,968MBA Purchase Index 311 1.4 6 2 61 325 186 265

Z-Score (12mo):Green 1.0 standard deviation low price or high yield/spreadYellow MeanRed 1.0 standard deviation high price or low yield/spread

52 Week

52 Week

Primary Market

MBS Snapshot

Change 52 Week

July 22, 2020

July 23, 2020 of 25


Z-Score Week MTD YTD High Low Avg

PACs

30 Year 2 yr 67 0.3 (5) (18) 25 108 39 603 yr 70 0.2 (5) (22) 22 114 45 664 yr 75 -0.1 (5) (20) 15 126 57 775 yr 80 -0.1 (5) (25) 13 133 64 837 yr 85 -0.2 (5) (25) 15 136 67 8910 yr 105 -0.2 (5) (20) 18 153 84 109

15 Year 2 yr 60 0.0 (5) (20) 18 108 39 603 yr 70 0.2 (5) (22) 22 114 45 664 yr 75 0.0 (5) (20) 15 126 57 755 yr 80 0.0 (5) (25) 15 131 62 807 yr 85 0.1 (5) (20) 20 131 62 8310 yr 105 0.0 (5) (15) 20 151 82 105

Sequentials

30 Year 2 yr 67 0.3 (5) (18) 25 108 39 603 yr 70 0.2 (5) (22) 22 114 45 664 yr 75 -0.1 (5) (20) 15 126 57 775 yr 80 -0.1 (5) (25) 13 133 64 837 yr 85 -0.2 (5) (25) 15 136 67 8910 yr 105 -0.2 (5) (20) 18 153 84 109

15 Year 2 yr 60 0.0 (5) (20) 18 108 39 603 yr 70 0.2 (5) (22) 22 114 45 674 yr 75 0.0 (5) (20) 15 126 57 755 yr 80 0.0 (5) (25) 15 131 62 807 yr 85 0.1 (5) (20) 20 131 62 8310 yr 105 0.0 (5) (15) 20 151 82 105

ARM (Z-spreads)

5x1 2/2/5 2.00 62 -0.6 (2) (2) 8 124 50 742.50 68 -0.6 (2) (2) 8 130 56 803.00 72 -0.6 (2) (2) 8 134 60 843.50 76 -0.6 (2) (2) 8 138 64 88

7x1 5/2/5 2.00 70 -0.6 (2) (2) 8 132 58 822.50 76 -0.6 (2) (2) 8 138 64 883.00 86 -0.6 (2) (2) 8 148 74 983.50 94 -0.6 (2) (2) 8 156 82 106

10x1 5/2/5 2.00 88 -0.5 (2) (2) 8 150 76 1002.50 94 -0.6 (2) (2) 8 156 82 1063.00 99 -0.6 (2) (2) 8 161 87 1113.50 101 -0.6 (2) (2) 8 163 89 113

* YTM** Spreads calculated to 15 CPB.

CMO Floater (Discount Margins)

Passthru 6.5 Cap 35 -1.9 0 (2) (6) 70 35 477.0 Cap 35 -1.6 0 (2) (2) 70 35 45

Support 5.0 Cap 80 -1.6 0 (5) (35) 125 80 935.5 Cap 75 -1.4 0 (5) (30) 120 75 876.0 Cap 70 -1.4 0 (5) (25) 115 70 83

CMO Spreads

July 22, 2020

Change 52 Week

July 23, 2020 of 25


Z-Score Week MTD YTD High Low AvgCMBS Spreads

New Issue 3y 100 0.5 (5) (5) 57 175 40 775y 105 0.2 (10) (10) 33 190 62 977y 115 0.1 (10) (10) 28 200 73 10910y 110 -0.2 (5) (5) 15 225 79 118

ACMBS

Fixed (N-Spread) 7y 40 -0.9 (1) (1) (7) 100 39 5010y 42 -1.3 (2) (2) (11) 115 42 57

Floating (DM) 7y 39 -1.2 (1) (1) (12) 70 39 4710y 45 -1.2 (2) (2) (12) 80 44 54

RMBS 2.0

AAA CC Price Drop 15yr (80.00) (110.06) 0.00 0.00 (79.50) (0.50) (2.25) (1.05)30yr (64.00) (87.10) 0.00 0.00 (63.06) (0.94) (2.75) (1.51)

Sprd to Swaps Front SEQ 125 n/a n/a n/a n/a n/a n/a n/a

Agencies

Bullets 2y 5.9 0.4 0.5 0.1 4.3 19.1 (0.5) 4.13y 11.4 0.9 1.3 0.5 6.7 25.9 1.5 6.85y 18.7 0.9 0.3 0.2 15.6 31.3 3.1 11.110y 33.6 0.4 2.8 1.2 11.9 63.8 18.4 28.4

Callables 5NC1 40.0 -0.2 2.4 (0.0) 11.1 95.8 27.0 43.37NC1 46.9 -0.8 0.9 (4.6) 4.2 111.7 38.9 59.010NC1 64.1 -1.2 (2.2) (1.9) (5.6) 134.1 64.1 83.815NC1 82.3 -0.9 (0.8) (0.1) 0.6 159.2 79.1 99.9

Week MTD YTD High Low Avg

Static

Price 107.07 1.3 (0.04) 0.09 2.66 107.43 103.30 105.42Coupon 3.43 -2.0 0.00 (0.03) (0.14) 3.67 3.43 3.55Yield 0.76 -1.7 (0.06) (0.18) (1.71) 2.84 0.71 1.92WAL 2.82 -1.7 (0.09) (0.17) (2.03) 6.25 2.77 4.12

Option-Adjusted

Effective Duration 1.28 -1.5 (0.05) (0.13) (1.77) 3.86 1.24 2.21Effective Convexity -0.16 1.9 0.04 0.22 1.59 -0.12 -2.10 -1.23LOAS (bps) 29 -1.2 (4.65) (12.56) (8.17) 153 14 47

Mix

30YR 89.8% -0.3 0.0% -0.2% -0.1% 90.0% 86.9% 89.9%15YR 10.2% 0.3 0.0% 0.2% 0.1% 13.1% 10.0% 10.1%

5 Day 0.07%10 Day 0.06%MTD -0.01%QTD 0.03%YTD 3.53%12 Month 5.19%

Source: MTGINDEX data from the Yield Book.

Nominal Return

MBS Index

Change 52 Week

Alternative Markets

Change 52 Week

July 22, 2020

July 23, 2020 of 25

This material was produced by an FHN Financial Strategist and is not considered research and is not a product of any research department. Strategists may provide information to investors as well as to FHN Financial’s trading desk. The trading desk may trade as principal in the products discussed in this material. Strategists may have consulted with the trading desk while preparing this material, and the trading desk may have accumulated positions in the securities or related derivatives products that are the subject of this material. Strategists receive compensation which may be based in part on the quality of their analysis, FHN Financial revenues, trading revenues, and competitive factors.

Some data in this report may be derived from information provided by CPR & CDR Technologies, Inc. Neither CPR & CDR Technologies, Inc. nor any of its directors, employees, or agents accept any liability for any loss or damage arising out of the use of all or any part of this report.

Although this information has been obtained from sources which we believe to be reliable, we do not guarantee its accuracy, and it may be incomplete or condensed. This is for informational purposes only and is not intended as an offer or solicitation with respect to the purchase or sale of any security. All herein listed securities are subject to availability and change in price. Past performance is not indicative of future results, and changes in any assumptions may have a material effect on projected results. Ratings on all securities are subject to change.

FHN Financial Capital Markets, FHN Financial Portfolio Advisors, and FHN Financial Municipal Advisors are divisions of First Horizon Bank. FHN Financial Securities Corp., FHN Financial Main Street Advisors, LLC, and FHN Financial Capital Assets Corp. are wholly owned subsidiaries of First Horizon Bank. FHN Financial Securities Corp. is a member of FINRA and SIPC — http://www.sipc.org.

FHN Financial Municipal Advisors is a registered municipal advisor. FHN Financial Portfolio Advisors is a portfolio manager operating under the trust powers of First Horizon Bank. FHN Financial Main Street Advisors, LLC is a registered investment advisor. None of the other FHN entities, including FHN Financial Capital Markets, FHN Financial Securities Corp., or FHN Financial Capital Assets Corp. are acting as your advisor, and none owe a fiduciary duty under the securities laws to you, any municipal entity, or any obligated person with respect to, among other things, the information and material contained in this communication. Instead, these FHN entities are acting for their own interests. You should discuss any information or material contained in this communication with any and all internal or external advisors and experts that you deem appropriate before acting on this information or material.

FHN Financial, through First Horizon Bank or its affiliates, offers investment products and services. Investment products are not FDIC insured, have no bank guarantee, and may lose value.

MORTGAGE STRATEGY MONTHLY

As of 7/22/2020

Payup 1-Month Carry B/E Libor Effective EffectiveCoupon Specification (ticks) Price WAC WALA Proj CPR ^ 1mo 3mo (ticks) Months YTM WAL OAS Duration Convexity

3.0 TBA (Cheapest to Deliver) 105.297 3.70 51 26.0 4.9 5.7 -3.94 1.12 3.04 55.94 1.62 -0.893.0 LLB 85k 128.0 109.297 3.51 45 14.8 7.4 7.5 -4.05 n/a 1.30 5.98 61.40 4.41 -0.913.0 MLB 110k 118.0 108.984 3.57 47 15.8 6.7 7.3 -4.18 n/a 1.21 5.44 51.87 3.73 -1.153.0 HLB 150k 91.0 108.141 3.51 46 17.0 5.9 6.8 -4.02 n/a 1.08 4.55 40.14 2.49 -1.543.0 175k Max 64.0 107.297 3.67 49 18.3 5.7 6.6 -3.82 528.9 1.07 4.07 43.21 2.14 -1.363.0 200k Max 38.0 106.484 3.67 48 20.8 4.9 6.0 -3.83 368.9 1.15 3.77 52.24 2.03 -1.313.0 New Wala 4.0 105.422 4.15 20 47.6 0.1 0.0 -9.40 n/a 0.50 2.28 12.82 0.90 0.343.0 20yr 10.0 105.609 3.61 47 28.2 5.1 5.6 -4.73 n/a 0.99 2.97 54.40 1.74 -0.423.0 Conv. Jumbo (CK) -52.0 103.672 3.77 48 43.6 3.8 4.0 -5.30 n/a 0.97 1.96 61.32 1.10 -0.283.0 100% Investor 36.0 106.422 3.92 51 21.1 4.9 4.9 -3.86 450.0 0.97 3.39 41.65 1.36 -1.41

3.5 TBA (Cheapest to Deliver) 105.422 4.07 40 33.4 7.3 8.0 -4.27 1.49 2.97 94.53 1.68 -0.633.5 LLB 85k 210.0 111.984 4.02 36 15.1 6.5 7.2 -4.23 5,675.7 1.14 5.50 48.77 3.57 -0.983.5 MLB 110k 186.0 111.234 4.05 39 18.6 6.9 7.4 -5.11 n/a 1.06 4.95 41.73 2.82 -1.083.5 HLB 150k 150.0 110.109 4.03 31 23.1 6.3 7.0 -5.95 n/a 0.88 4.10 28.48 1.71 -0.913.5 175k Max 110.0 108.859 3.96 50 24.8 6.1 6.5 -5.45 n/a 1.01 3.81 45.74 1.73 -0.873.5 200k Max 87.0 108.141 3.95 34 31.0 3.5 4.9 -6.77 n/a 0.90 3.33 38.84 1.08 -0.683.5 New Wala 16.0 105.922 4.61 19 66.2 2.7 2.7 -16.57 n/a -0.92 1.37 -87.55 -0.29 2.743.5 20yr 6.0 105.609 3.99 42 36.0 5.5 6.2 -5.12 n/a 1.32 2.78 89.71 1.72 -0.133.5 Conv. Jumbo (CK) -50.0 103.859 4.30 50 53.3 5.5 4.9 -6.31 n/a 1.56 2.21 114.31 1.83 0.133.5 100% Investor 52 107.047 4.18 51 28.0 5.6 6.3 -4.70 n/a 0.92 2.91 49.45 0.97 -0.23

4.0 TBA (Cheapest to Deliver) 106.297 4.48 36 35.1 6.6 7.1 -4.36 1.65 2.95 116.21 1.67 -0.294.0 LLB 85k 250.0 114.109 4.40 36 17.4 8.4 8.7 -5.06 n/a 1.07 5.17 45.29 3.11 -0.804.0 MLB 110k 210.0 112.859 4.39 34 23.3 7.4 8.3 -6.91 n/a 0.98 4.55 39.14 2.22 -0.864.0 HLB 150k 165.0 111.453 4.43 39 31.3 7.2 7.9 -9.24 n/a 0.78 3.76 26.81 1.45 -0.394.0 175k Max 120.0 110.047 4.38 50 24.7 7.0 7.7 -5.07 n/a 1.10 3.72 62.35 1.74 -0.444.0 200k Max 96.0 109.297 4.39 37 33.1 5.0 5.6 -7.45 n/a 0.75 3.02 37.74 0.87 0.104.0 New Wala 20.0 106.922 4.91 20 52.5 3.6 3.4 -11.33 n/a 0.77 2.27 46.46 1.48 0.714.0 100% Investor 52.0 107.922 4.64 41 36.6 7.2 7.5 -7.00 n/a 0.75 2.58 47.57 0.63 0.48

4.5 TBA (Cheapest to Deliver) 107.594 5.14 21 37.6 9.8 10.0 -5.45 1.46 2.72 108.24 1.40 0.314.5 LLB 85k 270.0 116.031 4.93 34 24.0 8.1 9.1 -8.50 n/a 1.16 5.19 57.04 3.17 -0.664.5 MLB 110k 227.0 114.688 4.88 42 24.3 9.2 9.4 -7.49 n/a 0.99 4.47 45.65 2.17 -0.554.5 HLB 150k 170.0 112.906 4.94 41 26.6 8.3 9.3 -7.08 n/a 1.06 4.02 57.76 1.82 -0.374.5 175k Max 120.0 111.344 4.84 54 29.4 8.0 9.0 -6.82 n/a 1.23 3.73 80.23 1.92 -0.154.5 200k Max 90.0 110.406 4.89 38 36.5 12.8 4.6 -8.88 n/a 0.99 3.16 64.43 1.46 0.294.5 New Wala 24.0 108.344 5.37 20 58.7 4.3 3.7 -16.32 n/a 0.62 2.26 32.71 1.61 0.754.5 100% Investor 44.0 108.969 4.84 50 30.6 8.8 9.4 -4.71 59.8 1.50 3.26 112.63 1.49 -0.01

Source: FHN Financial, Yieldbook

Specified Pool Carry and Breakevens

Cohort Hist. CPR

july 23, 2020 mtgage sategy my - fhn financial

Documents