We have been following American Capital Agency (AGNC) and Annaly Capital Management (NLY) for very nearly two years. These two companies are mortgage REITs, who are in the tricky business of borrowing money, using "leverage" and the derivatives markets to inflate it, and then lending it out to prospective homeowners, primarily with US Government backing. The reward for taking this risk is a nice, high dividend stream.
We have been tracking the two largest participants in this business over time to see what changes are going on, and also, to try to predict which of the two would be the better investment. The last time we did this was November, and there are links back to the previous articles for anyone who wants to see how we did on these predictions.
Here is the six month stock chart for these two stocks:
Although there has been a bit of a recovery in the last six weeks, it has generally been a rough period for investors in the mortgage REITs.
Here is a summary of the share price of these two stocks:
|AGNC||$ 31.14||$ 34.97||$ 30.82||$ 31.48|
|NLY||$ 16.29||$ 17.31||$ 14.98||$ 14.80|
More critically, here is a summary of the quarterly dividends:
|$/Share||Q4 2011||Q1 2012||Q2 2012||Q3 2012||Q4 2012|
|AGNC||$ 1.40||$ 1.25||$ 1.25||$ 1.25||$ 1.25|
|NLY||$ 0.57||$ 0.55||$ 0.55||$ 0.50||$ 0.45|
So you can see what happened: As we predicted a year ago, AGNC had the better time of it. The share price has been relatively stable, notwithstanding some nausea last November, and the stock has been able to maintain a stable $1.25 dividend. The competitor, NLY, in exactly the same business, was not able to do so.
There are a couple of reasons for this: Retail mortgage rates continued to decline all through the year last year, which put pressure on the margin that these companies make on their mortgages:
|Interest Rate Spread||Q1 2012||Q2 2012||Q3 2012||Q4 2012|
AGNC was able to maintain consistently higher mortgage rate spreads. Here is a comparison of leverage and also of the effects of the hedging programs of the two companies:
|Leverage||Q1 2012||Q2 2012||Q3 2012||Q4 2012|
The figures below are arrived at by summing the realized and unrealized derivatives and securities trading gains and losses from the quarterly reports:
|AGNC ($B)||$ 209||$ 129||$ 991||$ (242)|
|NLY($B)||$ 233||$ (765)||$ (307)||$ (976)|
AGNC was able to have a higher net interest spread because it expanded its portfolio more than NLY did via leverage, and also because it had a more successful hedging program, at least for the first nine months of the year.
So, how did these companies cope with this challenging environment of shrinking profit margins? Both companies issued series of preferred stocks which paid in the neighborhood of 7-8% dividends, in an effort to raise money at lower cost. Both companies also reduced their number of shares outstanding with common stock repurchases. Both of these activities have the effect of lowering the net cost of capital.
Annaly also took the unusual step of changing the company charter to allow investment of up to 25% of company equity in non-agency-backed securities, and announced the acquisition of Crexus (CXS) which was dealing in higher margin commercial mortgages. The jury is still out on that strategy, since it was supposed to take effect early this year. NLY's fourth quarter interest rate coupon value was 2.45% vs. 2.54% in the third quarter, according to the most recent announcement so the positive impact, if any, has not yet appeared.
Recently, AGNC announced an additional issuance of up to 57.5 million shares of additional common stock.
The net effect of the stock issuance is as follows, assuming the offering is priced near to the current market price.
|Est. Portfolio Increase||$ 12,135,375,000|
|Interest Spread (% Annual)||0.0144|
|Potential Margin $/year||$ 174,749,400 |
|Quarterly Dividend $/Share/Quarter||$ 1.25|
|Dividend Payout/Year||$ 287,500,000|
|Net Potential Effect of Issuance||$ (112,750,600)|
So, unless the management of AGNC knows something that we don't regarding the future direction of mortgage rates (or in anticipation of lower dividends), the strategy of increasing their portfolio with additional stockholder's equity appears to be a net negative.
Two other tables:
Here is the net book value per share of these two companies. This is a measurement used by "some" to determine whether either of the two funds is "cheap" as a function of their net stockholders' equity:
|Net Book Value per Common Share ($B)||Q1 2012||Q2 2012||Q3 2012||Q4 2012|
Currently, AGNC is selling at a bit of a premium to its book value, and NLY at a bit of a discount, but using this as a predictor of the future stock price has its flaws. If you had made your investment decision a year ago on that basis, you would have gone long on NLY and been worse off.
Here is the "Cash Position", the amount of cash on the books of each company: NLY raised cash in the third quarter to accomplish the share buyback and Crexus acquisition. Its cash amount has returned to a little less than its normal level. AGNC's cash position is still higher than it was in the first and second quarters.
|Cash Position ($B)||Q1 2012||Q2 2012||Q3 2012||Q4 2012|
|AGNC||$ 1,762||$ 2,099||$ 2,569||$ 2,430|
|NLY||$ 932||$ 924||$ 2,264||$ 615|
So, all of this being said, how much of the above discussion is a predictor of what is going to happen in the future?
According to this investors presentation the management of AGNC expects mortgage rates to remain low, but AGNC will benefit from improved financing because of a more predictable interest rate situation (the "dollar roll" market).
An additional key number is the fact that AGNC has gotten more conservative in the last quarter, reducing their leverage to 6.7 compared to their historical 8.0 at which it was leveraged during its growth phase in 2011.
A third key number is the net negative effect of AGNC's hedging program. We derived the numbers from the table above by this calculation: All values are in $M
|AGNC Q1||AGNC Q2||AGNC Q3||AGNC Q4|
|Net Interest Income||$408||$384||$381||$423|
|Gain on Sale of Securities||$216||$417||$210||$353|
|Loss on Derivatives (Net)||$47||$-1029||$-460||$89|
|Unrealized Gain/Loss on AFS Securities||$-106||$689||$1190||$-734|
|Unrealized Gain on Derivative Instruments||$52||$52||$51||$50|
|Net of Income and Hedging||$617||$513||$1372||$181|
|Net Effect of Hedging Program||$209||$129||$991||$-242|
Because of the $700M unrealized loss on available-for-sale securities, the AGNC hedging program did not insulate the company from the shrinking interest spreads like it did over the last year.
Based on these three graphs, I am hard pressed to believe that AGNC stock will continue to do better than NLY. However, NLY is faced with the same headwinds. The company's dividend has been in decline for the past year. The unknown number in NLY's case is the effect of the strategy change on its interest rate coupon, and we will have to wait a quarter or two to get that information.
Keep in mind what I always say: anywhere in the above calculation where there is an "if", "estimate", "might" or "assumption" there is an opportunity for error.
Also, keep in mind what I always say, that the world is full of chaos and there are no guarantees on anything.
It is probably time to look for an alternative to both of these mREIT giants, if one can be found that has a higher interest rate spread, while still maintaining some level of risk management. We have a few weeks before the next round of dividend announcements to look for some alternatives.
Additional disclosure: I went long on NLY shortly after my November article.