I have been skeptical to invest in the mortgage REIT space prior to the Fed tapering.  Seemingly, these names have no upside until rates stabilize at higher levels, which my gut says is at least 3.25% and more likely 3.5+% on the 10 year.  Indeed 10yrs traded at 3.5% prior to QE back in 2009.  However, the sheer cheapness of the space, and the wholesale selling of stocks with no regard to current durations or NAV offers opportunity today that is too good to pass up.  I think investors can make low risk, 20-30+% returns over the next 12-24 months, much in the form of cash dividends.  That is true even if rates end up at 3.5-4.0%.  

I encourage members to read the fine piece by rasputin998 on MFA a few months ago for a lot of good industry information.   Generally this year, mREITs have been killed owing to levered long positions in agency bonds.  Owning 106 dollar price agency bonds, levered 6-7x with prepays running at 20% seems insane in hindsight. Well, the shellacking to book value to almost every player has brought a new level of humility and understanding to management teams.  Nobody is bullish anymore on rates, and most have repositioned portfolios to lengthen liability duration substantially via swaps (and hence reduce net equity duration).  Leverage levels have also come down across the board.

There is a pervading fear in the space too, that once tapering really begins, the selloff in agency mortgage paper could be even worse.  Spreads on agency bonds to the 10 year have averaged roughly 100 basis points over time, and today we are seeing still somewhat tight, 80bps spreads.   As one mortgage trader told me however, the agency market is the 2^nd most liquid in the world.  Much of the tapering fear/liquidity selloff is already baked into many of these bonds. 

Generic Fannie 3.5s for example now trade under par, down from 106 and change just last May.  Another 20bps of spread widening might hit generic 30 years by another 1-1.5 points.  But portfolios are designed now to weather another 100bps of widening, with a shift away from long duration fixed 30 years, to 15 years, to ARMs, or away from agencies altogether.  NLY for instance has begun to invest in commercial RE paper, now representing 10% of their book.

Narrowing down to a specific name seemed a case of choosing 1) a good manager, 2) one trading at a decent discount to book, and 3) low duration so as to avoid capital losses as rates inevitably go higher. 

One would think that the names with the most leverage and duration risk would trade at the biggest discount to book.  That is true to an extent, but generally I have focused on which names have protected book value the best this year.  3 names stand out as ones with flat or slightly positive ROEs:  EFC, EARN, and MFA.  Here I am going with EARN, Ellington Residential.

Ellington Residential IPO’d at $20 in early Q2, so its better ROEs arguably could be due to its ramping up during Q2.  All of the pain occurred in Q2 in May and June after the company went public though, so I still give management a fair amount of credit here.  Recently too, Ellington Residential ’s management has authorized a 10mm share buyback, which is pretty meaningful for a $140mm market cap company. 

The buyback was announced in August, and as of November 13, no shares had been repurchased.   I have traded and watched many of these names over the past decade, and typically they trade from say 95% of book to 105% of book (rarely do you see buybacks).  Today the group hovers at about 80-85% of book, despite still positive (and growing) spreads, lower leverage levels and an improving housing market (which primarily helps the mREITs owning non-agency paper).  I don’t think the short end of the curve is likely to ding financing costs for the group for at least 2-3 more years either, maybe longer.

Ellington Residential’s sister company, EFC (Ellington Financial, not a REIT actually but a holder of RMBS), has been a solid performer owing to smart trading in the non-agency market over the past few years.  While admittedly Ellington Management charges hefty hedge fund sized fees (1.5 and 25 over 9), it’s hard not to see the value in the management team there.

The head honcho at Ellington Management Group (the external manager) is Mike Vranos.  He is quite well known as one of the smartest guys in mortgage land, and multiple people I have spoken with who know him or guys at his firm, say he is indeed one of the best.  He is the co-CIO, and EARN’s CEO is a top guy who has been at Ellington Management since 1995.  The firm’s overall 20 year track record bodes quite well here.  EFC has generated 12% avg annual ROEs since its founding in 2007, through some pretty tough times.  I expect over time EARN will be an outperformer too.

EARN’s buyback is exactly what EFC did last year, when that stock was trading at a big discount to BV.  It subsequently rallied to over book, at which point they did issue stock.  I encourage those interested to also read the MD&A sections of EFC, to get a sense for how these guys think and why they have been successful year in and year out.  EFC as one of the better performing names, trades at only a 5% discount to book.

Overall, EARN trades at a 16% discount to book, at 6.3x runrate core EPS (earnings based on interest, not capital gains), and only has roughly 5% exposure to rising rates.  That is, if rates rise 100 bps, then BV would fall 5% to around $17.40, meaning the stock would still trade at 89% of book.  The downside in 12 months seems limited.  I figure the upside could easily be a 1.0x Price/Book ratio if rates stabilize.  With dividends, that implies upside of 32% in a year.  Generally all of these names traded at book just 12 months ago.

Business

Ellington Residential’s focus is on Agency paper, including 15 and 30 year bonds.  The investing mandate is pretty broad though, and EARN can buy non-Agency paper, Mortgage Servicing Rights (MSRs), hybrids etc.  Generally EARN will focus on specified pools where they find value in Agency land.  That is, pools of perhaps low loan balance mortgages, or others with less likelihood of being prepaid.  In any case, their portfolio will shift to fit where there is value, and with a portfolio only $1.5BB of RMBS investments, the company can be quite nimble where others cannot.

Sensitivities To Rates / Upside To Stock

Overall, I see downside of 5% percent over a year if rates end up at 4% (post dividends), and the stock trades at a 25% discount to book.  I view this is a pretty draconian scenario.  Really the death knell for these names a large move in short term financing costs.  But with the fed promising low rates until 2016, it seems a low probability event, especially with inflation at such low levels.

Comps

These may be hard to read, but I ranked them in order of price to earnings (in green).  Here I took core earnings, that is net interest less fees, excluding capital gains and losses.  The good managers were able to protect book value this year.  EARN has essentially lost 1% of book (this includes gains/losses, total change in book after factoring in dividends).  This 1% is quite good, most managers are down in the double digits.  The only winners, EFC and MFA, had large credit books (ie non-Agency paper) which performed quite well, and less leverage. 

Other issues that investors need to think about. 

1) The SEC has occasionally discussed regulating these businesses.  My take is that there is a huge retail investor base, one that includes retirees and widows owning these stocks for yield.  When the SEC issued a Concept Release back in 2011 (one potentially proposing heavier regulation, less leverage, daily NAVs), it was clear that there was little political will to proceed and it went nowhere.  

2) Ellington Residential is more levered than some of their peers at 7.5x Debt/Equity.  Forced deleveraging could obviously have an impact here.  Haircuts may be raised by banks, although if memory serves, haircuts at 3% in 2008 were raised to 5%, not really impacting the public names.  A 5% haircut still enables 20x leverage, well above where anybody is levered today.

3) Convexity risk – if rates go up, then durations go up.  This “death spiral” fear has probably contributed to these names trading from book to big discounts to book.  Ellington manages their convexity with swaptions (options to enter into longer dated swaps if rates go higher), limiting this risk.  According to management, if rates rise 50bps, then book value / share would only be impacted by 40c taking it to $17.90. 

4) Q4 earnings – rates have moved up 30 bps.   I expect there to be further declines in book.  Figure 25c hit to book in EARN’s case.  Most names cut their dividend for Q4.  Dividends typically are an indication of earnings.  In Ellington’s case, they maintained their 50c dividend.  I also take some comfort in the fact that they are paying a dividend smaller their core EPS.  Other names with longer histories continue to use UTI’s (Undistributed Taxable Income) to pay larger dividends than what they are earning.  The Div Coverage column in the comp table above indicates that most of these names will continue to have to cut their distributions as UTI’s run out.

Conclusion

Ellington is a solid manager, and buying a bundle of agencies at a big discount to book seems to offer a pretty decent margin of safety.  There admittedly is no real moat in this business.  I view these guys as very good money managers though, and sometimes it is worth paying fees for outstanding performance.  While EARN has a short record and is a small cap name, over time I wouldn’t be surprised to see the company trade back to near book (or get to EFC’s 95% of book level), once the world is comfortable that rates have stabilized.  In fact, I give it decent odds that rates have already reached a steady level.  Below are 10 year government bond yields, illustrating that the US in the middle of the pack in terms of yield:

Will US yields sell off to levels similar to Spain, Italy and Mexico?  Seems unlikely.  Of course, the big issue with rates will be inflation.  Today’s 1.2% inflation rate implies 180 basis points of real return on the 10 year.  I still believe that short end stays low unless inflation really kicks in.

Here is CPI over the past 5 years:

Share buybacks

Rates reaching a level that makes investors comfortable that mREITs won’t continue to lose book value