MORTGAGE STRATEGY Monthly Report

Transcription

1 JANUARY 16, 219 MORTGAGE STRATEGY Monthly Report SOMA Impact on MBS P. 2 The FOMC surprised the market with a brief discussion about the potential to sell MBS assets out of SOMA in the minutes from the December 19 meeting that were released last Wednesday, January 9. Much of the commentary so far has been about the potential economic impact of such a move. This article will discuss the potential impact for MBS investors. Walt Schmidt Balanced Approach Wins the Day P. 7 As investors continue to decipher sometimes conflicting signals from various Fed speakers, we do not believe that now is the right time to make a large bet on yield curve shape. We continue to like trades combining 5-7 year bullet-like securities (such as FNMA DUS) with shorter bonds (such as Hybrid ARMs and CMOs). Tim Robinson MORTGAGE STRATEGIES Walt Schmidt, CFA walt.schmidt@ftnfinancial.com Tim Robinson, Ph.D tim.robinson@ftnfinancial.com Alexis Vilimas alexis.vilimas@ftnfinancial.com Market Update P. 1 Month-to-date, conventional 3yr and 15yr MBS have generally outperformed UST and swap hedge ratios. CMO spreads are 2-8bps wider since December while hybrid spreads are 2bps tighter. CMO Floater DMs were unchanged. The payups for loan balance specified pools 4.s-5.s increased 6-15 ticks since the end of last year. 1% NY payups also increased, by 5-12 ticks in 3.5s-5.s. Fixed prepayment speeds decreased in December by 5.8% due to elevated mortgage rates and slower seasonal turnover. Alexis Vilimas Outstanding Balance Billions Market Still on the Cusp of a Refi Event 3.s Marginally Refinanceable 5+bps ITM ~6.5% on 11/5/18 $217,19,637,226 ~17.8% on 1/15/19 $598,472,961,393 3yr Conventional Stack by WAC 3.5s 4.s 4.5s 5.s 5.5s 6.s Fully Refinanceable 1+bps ITM ~3.3% on 11/5/18 $111,54,769,593 ~6.6% on 1/15/19 $221,923,3,146 < > 7. Gross WAC Bucket Source: CPRCDR, Bloomberg, and FTN Financial Disclaimer is on the last page of this report. Willing and Able % FTNFINANCIAL.COM s Unwilling or Unable % 6% of balance 5bps in the money and "Willing and Able" 4.82% Note Rate on 11/5/ % Note Rate on 1/15/19

2 SOMA IMPACT ON MBS Executive Summary: The FOMC surprised the market with a brief discussion about the potential to sell MBS assets out of SOMA in the minutes from the December 19 meeting that were released last Wednesday, January 9. Much of the commentary so far has been about the potential economic impact of such a move. This article will discuss the potential impact for MBS investors. The modern MBS market has been in existence since Ginnie Mae issued its first public MBS in 197. Freddie Mac and Fannie Mae followed that in 1971 and 1981, respectively. The development of the CMO market followed in the mid-198s. However, the importance of the MBS market, and the GSEs combined impact on that market, increased substantially during the 199s. By 2, the combined Agency Fixed Rate MBS market had grown to just over $2 trillion and the GSEs (Freddie and Fannie combined) owned approximately 2% of that outstanding float. The GSEs had created a role for themselves as the marginal buyer of last resort that would keep a ceiling on MBS spreads. Since then, the size of the MBS market has more than tripled to $6.5 trillion (including Ginnie Mae securities), and the GSEs have handed the role of buyer of last resort to the Fed. A brief review of the portfolios of Freddie Mac and Fannie Mae since 2 are displayed in Figures 1 and 2 below. The portfolios are similar to each other in many ways, including size and general composition categories: own MBS/CMO, other agency paper, non-agency RMBS and mortgage loans. But there are also noticeable differences. Freddie retained portfolio focused on securities. Figure 1: Freddie Mac Retained Portfolio Has Had Greater Emphasis on Securities Unpaid Balance $ Bln 1, Freddie Mac Retained Portfolio Mortgage Loans Non-Agency Other Agency MBS/CMO Source: FTN Financial and GSE Filings January 16, 219 Page 2 of 14

3 SOMA IMPACT ON MBS Freddie Mac made a business decision in the middle of the previous decade to supplement its investments in Freddie MBS/CMO paper with a large non-agency RMBS book. By contrast, Fannie Mae had and continues to have a greater exposure to the loan book, both 1-4 family and multi-family. Fannie retained portfolio has more loan exposure. Figure 2: Fannie Mae Retained Portfolio Emphasizes the Loan Book Fannie Mae Retained Portfolio 1, Mortgage Loans 9 Non-Agency 8 Other Agency Unpaid Balance $ Bln MBS/CMO Source: FTN Financial and GSE Filings Fannie Mae did not break out mortgage loans from non-agency RMBS until 27, but its retained portfolio has had a higher percentage of investments in loans since 26. On a combined basis, the GSEs influence on the Agency MBS market reached a peak in terms of both actual holdings and percent of the outstanding market in late-23. This is clearly visible in Figures 3 and 4 below. Buyers of last resort in context of entire market. Figure 3: Outstanding Agency MBS Market by Portfolio Type Since 2 Holders of Agency Fixed MBS Outstanding Fixed MBS Market $ Trillions 7. Federal Reserve 6. Fannie Mae and Freddie Mac Combined Other Holders of Agency Fixed MBS Source: FTN Financial, CPRCDR, NY Fed, and GSE Filings January 16, 219 Page 3 of 14

4 SOMA IMPACT ON MBS This chart looks at three categories of investors in Agency MBS: the GSEs and the Fed, aka the buyers of last resort, and everyone else. The everyone else category grew impressively in and then again from 216-present. In both cases, US banks and foreign investors led the way. At the same time, the combined portfolios of the GSEs and the Fed have represented at least 15%, and in some cases more than 35%, of the entire ownership of the Agency MBS market since 2. GSEs and Fed dominant players in market over past 2 years. Figure 4: Combined Buyer of Last Resort Ownership of MBS Market 4% GSE and Fed Ownership of Agency Fixed MBS Market % Ownership of Outstanding Market 35% 3% 25% 2% 15% 1% 5% Fannie Mae and Freddie Mac Retained Portfolio Fed MBS Holdings % Source: FTN Financial, CPRCDR, NY Fed, and GSE Filings Things can change quickly in the market. For example, the percentage of the MBS market that was controlled by the GSE portfolios fell from 35% in late 23 to only 15% less than five years later in early 28. Shortly thereafter, the Fed embarked on a dramatic increase in the SOMA balance sheet in QE1 that pushed the ownership percentage of the combined Fed and GSE portfolios to over 35% of the market. After a lull in MBS purchases by the Fed in 21 and 211 that saw the combined ownership percentage drop quickly to 25% of the market, it re-embarked on MBS purchases in the form of QE3 that pushed the ownership percentage back over the 35% mark. Since then, the combined ownership percentage has slowly decreased to just below 3%, where it is presently. This is where the recent discussion about the SOMA balance, especially as it relates to the MBS portion, comes into play. The FOMC has already stated its intention to reduce the balance sheet to a size less than during the crisis (approximately $4.5 trillion) but greater than the pre-crisis size of approximately $9 billion. It has also stated its intention to eventually reduce its holdings of MBS to zero. We would be very concerned about current MBS valuations if the Fed were to speed up that process within the foreseeable future. Here is why. January 16, 219 Page 4 of 14

5 SOMA IMPACT ON MBS Buyer of last resort impact on OAS levels. Figure 5: Inverse Relationship Between Ownership % and OAS 3-Year Fannie MBS OAS vs. % of Market Owned by Fed and GSEs 35 4% GSE % Ownership 3 Fed % Ownership 35% 25 3yr OAS 3% OAS (bps) % 2% 15% % Ownership 5 1% 5% -5 % Source: FTN Financial, CPRCDR, NY Fed, YieldBook, and GSE Filings There are, of course, many factors that play into OAS levels for MBS product. With that in mind, we find it more than a coincidence that over the past 19 years, there seems to be a fairly strong inverse relationship between the % of the market that is controlled by the buyer of last resort and MBS OAS levels. The chart in Figure 6 more clearly displays this inverse relationship. The fit line is not perfect, but the implication is clear. Figure 6: Less Involvement from the Buyer of Last Resort May Be Problematic 3yr OAS Year Fannie TOAS vs. % Owned by GSEs and Fed Citi MBS Index Data Pre-Crisis Post-Crisis -2 Crisis Predicted 3yr OAS % 2% 25% 3% 35% 4% % Combined Ownership Y= X + e R-Squared: 56.9% Source: FTN Financial, CPRCDR, NY Fed, YieldBook, and GSE Filings January 16, 219 Page 5 of 14

6 SOMA IMPACT ON MBS This is the bottom line result to the monthly data that we have collected over the past almost 2 years. The fit line is not perfect, but we are fairly confident to say that given current supply/demand dynamics a further reduction in the Fed s portfolio will likely lead to wider OAS levels over time. But we are on record as having a positive outlook on the MBS basis for 219 absent a major refi event. That seems incongruous with these data. The increased demand from both domestic banks and foreign investors should continue to be relatively good in 1H19 barring any catastrophic policy errors with the hand-off of the FHFA to a new Director. Furthermore, the FTN Financial Interest Rate Strategies Group anticipates that Treasury supply especially in notes and bonds will be higher in 219 by about 3% YOY ($19 billion) and 15% relative to the 4Q18 run rate. On the other hand, we expect net MBS supply to $25 billion vs. $335 billion in 218. So, this leaves the Fed in the spotlight. Based on current prepayment assumptions, the MBS portion of the Fed s MBS holdings is expected to naturally amortize by $185 billion, and there should be $262 billion of Treasury maturities in the SOMA portfolio in 219. Therefore, both supply functions both actual projected and that implied by SOMA roll-off favor MBS over Treasuries in the near term. However, it is difficult to escape the reality of Figure 6 above. There are three main factors to any market: supply, demand and underlying risk factors. If the Fed changes course and decides to unwind MBS at a faster clip than simply allowing the MBS portfolio to roll off, that will affect both demand AND add a risk factor that the market is not anticipating. If that were to happen, we would likely change our current overweight posture directly to underweight without stopping at neutral. The MBS market has a good thing going right now. Spreads, expected returns AND underlying mortgage rates for borrowers are all reasonable at current levels. The implementation of Single Security and the Freddie to UMBS exchange will add plenty of angst and uncertainty to the market over the next few months. If the Fed adds more uncertainty through MBS asset sales, it could spell trouble for MBS valuations going forward. January 16, 219 Page 6 of 14

7 BALANCED APPROACH WINS THE DAY Executive Summary: As investors continue to decipher sometimes conflicting signals from various Fed speakers, we do not believe that now is the right time to make a large bet on yield curve shape. We continue to like trades combining 5-7 year bullet-like securities (such as FNMA DUS) with shorter bonds (such as Hybrid ARMs and CMOs). Figure 1 shows the difference in basis points among three points along the UST yield curve. The 2/1 and 5/1 spreads have actually widened since early December. The 2/5 spread has recently been drifting above and below zero. If the long end of the curve comes down, bullet securities like FNMA DUS will perform tremendously well. But if the Fed continues to push the short end of the curve higher, bonds like seasoned Hybrid ARMs and short CMOs will outperform. We believe that the right strategy in the current environment is a combination of these complementary security types. 2/5 Spread hovering around zero. Figure 1: Curve Shape 8 Steepness of UST Curve Spread (bps) to 5 Spread 5 to 1 Spread 2 to 1 Spread Source: FTN Financial, Bloomberg Figure 2 shows Hybrid ARM Z spreads since the beginning of 218. Z spreads were in the teens for much of the first quarter of 218. Current Z spreads are in the mid to upper 3s. Compelling value is available in the Hybrid ARM sector. January 16, 219 Page 7 of 14

8 BALANCED APPROACH WINS THE DAY Hybrid ARMs Z spreads have widened. Figure 2: Hybrid ARM Z Spreads Z Spread (bps) Conventional 5x1 2.5 Coupon Z Spread Source: FTN Financial Combine complementary bonds to outperform. Figure 3 shows a trade where an investor buys a 4%/6% combination of a seasoned 5x1 (3 months to the first reset) and a 7/5 DUS. The Hybrid ARM coupon can increase by as much as 2 basis points in 3 months. The Hybrid/DUS combination increases yield by 19 basis points and wins handily in 4 of the 7 total return scenarios. The trade wins modestly in the base case and essentially ties in the 2 flattening scenarios. Figure 3: Hybrid ARM/DUS Combination Security Analysis as of 1/16/219 Note: CPY assumed for DUS in all scenarios Bond Current Avg Libor Eff Eff # Security Type Wght Price WALA WAM Coupon WAC YTM Life OAS Dura Convx BUY FH Seas 5x1 4% BUY FN DUS 6% Combination 1% SELL FNCI yr NA Total Return Difference 12mo TRR Difference (bps) dnsteep dnflat dn1 base up1 upflat upsteep Rate Shock: 12mo Horizon (Gradual Shock) All analytics performed on the Yield Book using the current prepayment model. Total Return analytics assume constant-oas pricing. upsteep - 3moL up 25bps, 5yr swap up 5bps, 1yr swap up 75bps upflat - 3moL up 75bps, 5yr swap up 5bps, 1yr swap up 25bps dnflat - 3moL dn 25bps, 5yr swap dn 5bps, 1yr swap dn 75bps dnsteep - 3moL dn 75bps, 5yr swap dn 5bps, 1yr swap dn 25bps Source: FTN Financial, YieldBook January 16, 219 Page 8 of 14

9 BALANCED APPROACH WINS THE DAY Figure 4 shows a similar trade as Figure 3, but here a CMO is utilized in place of the Hybrid ARM. The CMO is about a 3yr in the base case (3.38yr Bloomberg median, 2.96yr Yield Book) and extends to just under 5.8yr up3 (Bloomberg median). The trade improves yield by 28 basis points and wins in 6 of the 7 total return scenarios and essentially ties in the 7 th. Short CMO/DUS combination provides value. Figure 4: CMO/DUS Combination Security Analysis as of 1/16/219 Note: CPY assumed for DUS in all scenarios Bond Current Avg Libor Eff Eff # Security Type Wght Price WALA WAM Coupon WAC YTM Life OAS Dura Convx BUY FNR DA CMO 32% BUY FN DUS 68% Combination 1% SELL FNCI yr NA Total Return Difference 12mo TRR Difference (bps) dnsteep dnflat dn1 base up1 upflat upsteep Rate Shock: 12mo Horizon (Gradual Shock) All analytics performed on the Yield Book using the current prepayment model. Total Return analytics assume constant-oas pricing. upsteep - 3moL up 25bps, 5yr swap up 5bps, 1yr swap up 75bps upflat - 3moL up 75bps, 5yr swap up 5bps, 1yr swap up 25bps dnflat - 3moL dn 25bps, 5yr swap dn 5bps, 1yr swap dn 75bps dnsteep - 3moL dn 75bps, 5yr swap dn 5bps, 1yr swap dn 25bps Source: FTN Financial, YieldBook As we all continue to digest sometimes conflicting signals from various Fed speakers, a balanced portfolio strategy is prudent. Combining complementary securities will produce outperformance in a range of interest rate scenarios. Attractive value is available in short bonds such as Hybrid ARMs and CMOs, while 7/5 DUS bonds provide generous spreads in the 5-7yr part of the curve. Combining a short CMO or Hybrid ARM with a DUS bond improves yield and total return performance. January 16, 219 Page 9 of 14

10 MARKET UPDATE MBS Snapshot Change 52 Week January 15, 219 Z-Score* Week MTD YTD High Low Avg Prices 3 Year (.28) (.6) (.6) (.25) (.6) (.6) (.16) (.2) (.2) (.8) (.2) Year (.16) (.27) (.14) (.14) (.31) (1.92) (3.41) Year (.2) (.25) (.16) (.16) (.14) (.6) (.6) (.8) I-Spreads (UST) 3 Year (1.1) (.7) (1.3) (1.2) (2.3) (3.1) (3.8) (6.1) (3.7) (9.6) (4.5) (12.6) (6.7) (11.4) (6.3) (9.4) Year (4.2) (1.4) (2.1) (1.5) (3.8) (1.8) Primary Market (6.8) (.8) Mortgage Rates 52 Week Conforming 3 Year (.6) (.8) (.8) Year (.1) (.1) x1 Hybrid Borrower Activity 52 Week MBA Refinance Index , ,9 MBA Purchase Index Z-Score (12mo): Green 1. standard deviation low price or high yield/spread Yellow Mean Red 1. standard deviation high price or low yield/spread January 16, 219 Page 1 of 14

11 MARKET UPDATE CMO Spreads Change 52 Week January 15, 219 Z-Score Week MTD YTD High Low Avg PACs 3 Year 2 yr yr yr yr yr yr Year 2 yr yr yr yr yr yr Sequentials 3 Year 2 yr yr yr yr yr yr Year 2 yr yr yr yr yr yr (3) (3) (3) ARM (Z-spreads) 1x1 (CMT) n/a n/a n/a 1x1 (LIBOR) n/a n/a n/a 5x1 2/2/ (2) (2) (2) (2) (2) (2) (2) (2) (2) x1 5/2/ (2) (2) (2) (2) (2) (2) (2) (2) (2) x1 5/2/ (2) (2) (2) (2) (2) (2) (2) (2) (2) * YTM ** Spreads calculated to 15 CPB. CMO Floater (Discount Margins) Passthru 6.5 Cap Cap Support 5. Cap Cap Cap January 16, 219 Page 11 of 14

12 MARKET UPDATE Alternative Markets Change 52 Week CMBS Spreads January 15, 219 Z-Score Week MTD YTD High Low Avg New Issue 3y (2) (2) (2) y (5) (5) (5) y (5) (5) (5) y (5) (5) (5) ACMBS Fixed (N-Spread) 7y (4) (4) (4) y (4) (4) (4) Floating (DM) 7y (1) (1) (1) y (1) (1) (1) RMBS 2. AAA CC Price Drop 15yr (1.13) n/a.. n/a n/a n/a n/a 3yr (1.75) (1.77)... (.84) (1.75) (1.26) Sprd to Sw aps Front SEQ 9 n/a n/a n/a n/a n/a Agencies Bullets 2y (1.8) (2.5) (2.5) y y (.9) y (1.1) (.5) (.5) Callables 5NC (5.2) (2.1) (2.1) NC (8.1) (4.7) (4.7) NC (3.6) (2.1) (2.1) NC (3.1) (2.4) (2.4) MBS Index Change 52 Week Week MTD YTD High Low Avg Static Price (.8) Coupon Yield (.2) (.2) WAL (.2) (.2) Option-Adjusted Effective Duration Effective Convexity (.2) (.2) LOAS (bps) (.5) (.95) (.95) Mix 3YR 89.2% 2.1.%.1%.1% 89.2% 87.9% 88.4% 15YR 1.8% -2.1.% -.1% -.1% 12.1% 1.8% 11.6% Nominal Return 5 Day -.1% 1 Day.56% MTD.22% QTD.22% YTD.22% 12 Month 1.51% Source: MTGINDEX data from the Yield Book. January 16, 219 Page 12 of 14

13 MARKET UPDATE Specified Pool Carry and Breakevens As of 1/15/219 Payup 1-Month Cohort Hist. CPR Carry B/E Libor Effective Effective Coupon Specification (ticks) Price WAC WALA Proj CPR ^ 1mo 3mo (ticks) Months YTM WAL OAS Duration Convexity 3. TBA (Cheapest to Deliver) LLB 85k n/a MLB 11k n/a HLB 15k n/a k Max New Wala n/a yr n/a Conv. Jumbo (CK) n/a % Investor n/a TBA (Cheapest to Deliver) LLB 85k n/a MLB 11k n/a HLB 15k n/a k Max n/a New Wala n/a yr n/a Conv. Jumbo (CK) n/a % Investor n/a TBA (Cheapest to Deliver) LLB 85k n/a MLB 11k n/a HLB 15k n/a k Max n/a New Wala yr n/a Conv. Jumbo (CK) n/a % Investor TBA (Cheapest to Deliver) LLB 85k n/a MLB 11k n/a HLB 15k n/a k Max n/a New Wala n/a yr Conv. Jumbo (CK) n/a % Investor TBA (Cheapest to Deliver) LLB 85k n/a MLB 11k n/a HLB 15k n/a k Max n/a New Wala n/a % Investor January 16, 219 Page 13 of 14

14 This material was produced by an FTN 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 FTN 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, FTN Financial revenues, trading revenues, and competitive factors. Some data in this report may be derived from information provided by CPR & CDR Technology, Inc. Neither CPR & CDR Technology, 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. Ratings on all securities are subject to change. Opinions, historical price(s) or value(s) are as of the date and, if applicable, time, indicated. FTN Financial does not accept any responsibility to update any opinions or other information contained in this communication. FTN Financial is not providing investment advice through this material. This is for information purposes only and is not intended as an offer or solicitation of any product. Securities, financial instruments, products or strategies mentioned in this material may not be suitable for all investors. Before acting on any information in this material you should consider whether it is suitable for your particular circumstances. Further information on any of the securities or financial instruments mentioned in this material may be obtained upon request. FTN Financial Group, FTN Financial Capital Markets, FTN Financial Portfolio Advisors and FTN Financial Municipal Advisors are divisions of First Tennessee Bank National Association (FTB). FTN Financial Securities Corp (FTSC), FTN Financial Main Street Advisors, LLC, and FTN Financial Capital Assets Corporation are wholly owned subsidiaries of FTB. FTSC is a member of FINRA and SIPC FTN Financial Municipal Advisors is a registered municipal advisor. FTN Financial Portfolio Advisors is a portfolio manager operating under the trust powers of FTB. FTN Financial Main Street Advisors, LLC is a registered investment advisor. None of the other FTN entities including, FTN Financial Group, FTN Financial Capital Markets, FTN Financial Securities Corp or FTN Financial Capital Assets Corporation 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 FTN 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. FTN Financial Group, through FTB or its affiliates, offers investment products and services. Investment Products are not FDIC insured, have no bank guarantee and may lose value. January 16, 219 Page 14 of 14