The mREIT stocks have relatively few distinguishing characteristics. Structurally, they pay out 90% of their taxable income via dividends. In order to do so, they utilize leverage ranging from 4x to 11x, in order to fund assets currently earning 2% to 4% at financing rates ranging from 1% to 2%.
The two most prominent stocks by both market cap and trading volume are NLY and AGNC. Earlier this year, AGNC transformed itself via an acquisition that internalized its management and reduced its pro-forma run-rate operating expenses to less than 100 bps.
Accompanying this transaction, AGNC released a slide deck that showed the drag from management fees within the sector. This a short idea for Armour Residential REIT, Inc (ticker: ARR) whose operating cost structure at 371 bps are significantly higher than the sector’s bellwether stocks of AGNC and NLY at a respective 70 bps (inclusive of MTGE fees) and 152 bps. This should prove to be an ever present impediment to value creation for shareholders. In the presentation slide below, we believe that ARR is “REIT E”:
Unfortunately, REITs A and B aren’t good comps, and our strategies include trading pairs of close substitutes. (In full disclosure, we are often buying, selling or shorting any of these given stocks in our capacity as investors and traders. Please see additional disclaimer appended below).
In addition, ARR’s fee structure reveals flaws in its original incentives that has been compounded by a regrettable track record. ARR pays a management fee equal to 1.5% of gross equity raised on their first $1B and then 0.75% of gross equity raised in excess of $1B. To date, ARR has raised gross equity of nearly $2.5B but has only $1.2B of equity remaining on their balance sheet. Shareholders are stuck paying a management fee based on that $2,469,368,000 while over half that equity no longer seems to need very much managing. Given how ARR structured its management fee, the only way this expense would decrease is by repurchasing shares or paying dividends in excess of taxable income. While ARR does have a repurchase program, and to their credit repurchased shares aggressively in Q4 last year (at approximately 73% p/nav), as of November 1st, they have reported no repurchases in 2016.
Here is the melting ice cube of ARR’s NAV/share, which is split adjusted for ARR’s 1 for 8 reverse stock split on 8/3/15 (and keep in mind that according to ARR, their NAV has fallen an additional 9.8% from 9/30th to 11/8th):
The concept of a management fee that more closely tracks the high water mark of equity issuance rather than the actual equity under management is interesting to say the least:
This is not a complicated thesis. Mortgage REIT’s are a retail oriented investment product that exhibit a strong tendency to converge toward similar dividend yields. A quick view of today’s current prices and dividend yields:
We utilized two steps above to come up with current P/NAV ratios. We begin with the most recent reported NAV updates of AGNC and ARR, and then adjust all three securities by their historical gains & losses correlations to each other. Because AGNC currently has very similar interest rate sensitivities as NLY and a strong historical correlation, we do a stepwise approach more dependent on AGNC to estimate NLY’s NAV. The goal here is not precision but relative value description for trading purposes.
They obvious question is whether or not all 11.6%ish dividend yields are comparably the same? Utilizing what we’ve calculated for operating expenses, an interesting thought experiment would be to compare the “net” dividend yield at which the stocks are currently trading versus the “gross” dividend yield that can be backed into given their reported operating expenses:
Dvd Yld on NAV (note 1):
Pre-OpEx Dvd Yld on NAV:
Note 1) based on our estimated NAV’s of ARR = $25.15, AGNC = $21.86, and NLY = $11.84
Is it possible that Armour shares, risk-adjusted, should trade at the same dividend yield as their more successful mREIT peers and comparable P/NAV despite the significant drag from their operating expense ratio? Or does the relative differences in dividend yields net of operating expense reflect that shareholders are embracing greater risk to cover the fee structure for a comparable product? The table above suggests the trade-off necessary to get to ARR’s comparable dividend yield.
Armour’s operating characteristics are very straightforward (as of 11/18/16 from company presentation dated 11/28/16):
Cost of Funds
Net Interest Margin
Common stock price: $22.18 versus Estimated book value $25.13 for 0.883x P/NAV as of 11/18th
Asset Balance Sheet Duration: 4.15
Hedge Balance Sheet Duration: (2.15)
Net Balance Sheet Duration: 1.99
GAAP Leverage (which excludes TBA dollar rolls or forward settling transactions): 7.2x
That last leverage calculation doesn’t reflect leverage as Armour and most mREITs report it quarterly. Adjusted for ARR’s 11/18th reported use of TBA’s (slide 7) and adjusted repo balance for forward settling trades (slide 11), leverage is approximately 8.5x.
Finally, ARR’s portfolio composition on both the asset and liability side show a strong similarity to AGNC and NLY. Last quarter’s reported numbers:
Cost of Funds
Net Interest Margin
In essence, ARR is burdened by the strong headwind of its fee structure. This is masked by the surface valuations that drive share prices in this retail dominated sector of the market.
There is one caveat, however, as sometimes bad is good, and in this case, it relates to the historical losses built up by ARR. Net capital losses realized in 2013, 2014 and 2015 combine to net over $900M and will be available to offset future capital gains:
This would be more worrisome to the short case if Armour had exhibited a track record of significant capital gains.
Armour is currently rich relative to its closest peers given the drag of its fee structure, and in particular it seems expensive relative to the ever improving cost structure at AGNC.
We estimate that on relative value basis, AGNC is arguably 800 bps cheap to ARR and has more compelling characteristics, which is another way of saying that even at fair value we’d rather stay in that spread than face the fee hurdle. Perhaps at twice that discount, which if it happened overnight would roughly correspond to where ARR management has historically bought back shares on an absolute p/nav basis, would ARR prove attractive in spite of its fee structure.
That is because while Armour’s management fee structure is a disadvantage, ARR did aggressively repurchase stock in Q3 and especially Q4 last year, reducing the share count from 43.8 million shares to 36.7 million shares, a reduction of 16.2%. Very roughly, we estimate that ARR bought back a considerable amount of these shares at approximately a 73% P/NAV.
We understand that this is the Value Investors Club, not the Relative Value Investors Club (RVIC…feel free to run with that idea anyone âº), but as both a trading strategy and a source of short ideas, close substitutes with distinct operating expense differences can be a reliable source of future investing and hedging strategies.
The author of this posting and related persons or entities (“Author”) currently has a financial stake in this security. Author may buy or sell or short additional shares, or adjust some or all of Author’s position, at any time. Author has no obligation to inform anyone of any changes to Author’s view. Please consult your financial, legal, and/or tax advisors before making any investment decisions. While the Author has tried to present facts it believes are accurate, the Author makes no representation as to the accuracy or completeness of any information contained in this note. The reader agrees not to invest based on this note, and to perform his or her own due diligence and research before taking a position in any stocks mentioned above. READER AGREES TO HOLD AUTHOR HARMLESS AND HEREBY WAIVES ANY CAUSES OF ACTION AGAINST AUTHOR RELATED TO THE NOTE ABOVE. As with all investments, caveat emptor.
I do not hold a position with the issuer such as employment, directorship, or consultancy. I and/or others I advise hold a material investment in the issuer's securities.
Fee structures continue to be the dominant trend in differentiating investment vehicles.
Continued poor relative NAV performance to its peers relative to its fee structure.