American Capital Agency's Dividend Projection For February 2016 - March 2016

| About: AGNC Investment (AGNC)

Summary

This article analyzes AGNC’s near-term dividend sustainability by performing two tests based on recent quarterly results. This article also discusses some current and future factors to consider.

Test 1 analyzes AGNC’s quarterly estimated REIT taxable income and net dollar roll metrics, while Test 2 analyzes the company’s recent cumulative undistributed taxable income and net dollar roll metrics.

My exact AGNC dividend per share projection for February 2016 – March 2016 (and corresponding probabilities) is stated in the “Conclusions Drawn” section of the article.

My exact MTGE and NLY dividend per share projections for the first quarter of 2016 are stated near the end of the article.

My current buy, sell, or hold recommendation for AGNC, MTGE, and NLY are stated in the “Conclusions Drawn” section of the article.

Author's Note: This article provides a detailed analysis with supporting documentation on the "most probable" dividend per share rate American Capital Agency Corp (NASDAQ:AGNC) will declare for February 2016 - March 2016. I perform this type of detailed analysis for readers who anticipate/want such an analysis performed at periodic intervals. For readers who just want the summarized conclusions/results, I would suggest to scroll down to the "Conclusions Drawn" section at the bottom of the article.

Focus of Article:

The focus of this article is to provide a detailed analysis with supporting documentation (via two tests) on the near-term dividend sustainability of AGNC, including a dividend per share projection for February 2016 - March 2016. I am writing this particular article due to the continued high demand that such an analysis be performed, especially in light of recent quarterly earnings and events. Understanding the tax and dividend payout characteristics of AGNC will provide investors with an overall better understanding of the mortgage real estate investment trust (mREIT) sector as a whole. Due to the fact AGNC has produced an annual dividend yield of at least 10.5% since the company's inception in 2008, many investors have chosen this stock (including other stocks within the mREIT sector) for income-producing equity investments. From reading this article, investors will better understand how a qualified real estate investment trust ("REIT") per the Internal Revenue Code ("IRC") comes up with an entity's current dividend rate and specific signs when an impending dividend raise or cut should be implemented.

I will be performing two dividend sustainability/projection tests within this article. These two tests will be termed "TEST 1" and "TEST 2". This will also include a brief analysis of AGNC's affiliate American Capital Mortgage Corp. (NASDAQ:MTGE) and a discussion of the company's closest sector peer Annaly Capital Management, Inc. (NYSE:NLY) regarding the same general metrics. At the end of this article, there will be a conclusion regarding my personal projection of what I believe AGNC's dividend per share rate will be for February 2016 - March 2016.

Side Note: I recently wrote a three-part article on AGNC where I projected the company's income statement (also known as the consolidated statement of comprehensive income) for the fourth quarter of 2015. In a second article, I projected AGNC's book value ("BV") as of 12/31/2015. These two sets of articles set the foundation for understanding certain types of account descriptions and figures shown in several tables within this article's dividend sustainability and projection analysis. As such, I would recommend new readers to first look at my previous articles to gain a better understanding of AGNC's accounts and how certain figures are derived. The following links are to my recent AGNC income statement, BV, and quarterly assessment articles:

American Capital Agency Corp.'s Q4 2015 Income Statement Projection - Part 3

American Capital Agency Corp.'s Upcoming Q4 2015 Book Value Projection

Assessing American Capital Agency Corp.'s Results For Q4 2015

Estimated REIT Taxable Income ("ERTI")/ERTI and Net Dollar Roll Overview:

Before we begin AGNC's dividend sustainability and projection analysis, let us briefly get accustomed to the information provided in Table 1 below. This table shows AGNC's ERTI/ERTI and net dollar roll from the fourth quarter of 2015 going back to the first quarter of 2015. I also include my projected figures for the fourth quarter of 2015 for comparative purposes. All figures within Table 1 are for the "three-months ended" (quarterly) timeframe.

Table 1 - AGNC Quarterly ERTI/ERTI and Net Dollar Roll Analysis

Click to enlarge

(Source: Table created entirely by myself, partially using data obtained from quarterly investor presentation slides)

Due to the specific IRC provision stating an entity must distribute at least 90% of its annual REIT taxable income ("AREITTI") to retain the company's qualified REIT status, AGNC usually bases the company's current and future dividend per share rates off of quarterly ERTI. However, due to a specific scenario that has unfolded regarding AGNC's "true earnings power" and "capital loss carryforward" balance, the company's quarterly ERTI/AREITTI is currently not the "best" figure to base its dividend distributions off of. Currently, as management has implied, I believe the best metric to use is AGNC's quarterly ERTI and net dollar roll figure. With that being said, quarterly ERTI is still an important indicator regarding MINIMUM annual distribution requirements ("ADR"). Now let us perform AGNC's dividend sustainability and projection analysis.

TEST 1 - Quarterly ERTI/ERTI and Net Dollar Roll - Common Shareholders Versus Quarterly Distributions Analysis:

Before we begin TEST 1 of AGNC's dividend sustainability and projection analysis, let us first briefly get accustomed to the information provided in Table 2 below. Table 2 is an extension of the information provided in Table 1 above. Table 2 below shows AGNC's ERTI/ERTI and net dollar roll from the fourth quarter of 2015 going back to the first quarter of 2015. I also include my projected figures for the fourth quarter of 2015 for comparative purposes. All figures within Table 2 below are for the "three-months ended" (quarterly) timeframe. Table 2 compares AGNC's quarterly ERTI/ERTI and net dollar roll figures to the company's dividend distributions figure showing the quarterly underpayment (overpayment).

Table 2 - AGNC Quarterly ERTI/ERTI and Net Dollar Roll - Common Shareholders Versus Quarterly Distributions Analysis (TEST 1)

Click to enlarge

(Source: Table created entirely by myself, partially using AGNC data obtained from the SEC's EDGAR Database)

AGNC does not provide a table that is comparable to most of Table 2 above. As such, there will not be an identical sheet AGNC provides that matches the data I have prepared in Table 2. I have gathered specific information derived from multiple tables/charts for a more detailed analysis.

TEST 1 - Analysis and Results:

Using Table 2 above as a reference, I take AGNC's quarterly "ERTI - common shareholders" figure (see red reference "E" in Table 2 above) and subtract this amount by the quarterly "distributions to shareholders from ERTI" figure (see red reference "I" in Table 2 above). If AGNC's red reference "E" is greater than the company's red reference "I", then AGNC technically had enough quarterly ERTI to pay out the company's dividend distributions for a particular quarter. As such, any excess quarterly ERTI left over (after accounting for the dividend distributions) would be added to AGNC's cumulative "undistributed taxable income" ("UTI") balance. UTI will be further discussed within TEST 2 later in the article. If AGNC's red reference "E" is less than the company's red reference "I", then AGNC had overpaid the company's quarterly dividend distributions and must use a portion of the remaining cumulative UTI balance (or add to the deficit balance) to help pay for the overpayment.

AGNC reported quarterly ERTI of $91 million or $0.27 per common share for the fourth quarter of 2015. Even though this amount was modestly higher when compared to quarterly ERTI of only $59 million or $0.17 per common share for the third quarter of 2015, it should be noted this amount still remained at suppressed levels. Let us discuss one main reason why AGNC's quarterly ERTI has remained well below the company's historical average during the prior several quarters. In order to explain this, let us take a step back and understand how this figure was derived in prior quarters.

Throughout 2014, AGNC continued to maintain a material "off-balance sheet" net long "to-be-announced" ("TBA") MBS position while refraining from increasing the company's "on-balance sheet" regular MBS portfolio. As of 3/31/2014, 6/30/2014, 9/30/2014, and 12/31/2014, AGNC had a net long TBA MBS position of $13.9, $17.8, $17.3, and $14.4 billion, respectively (based on notional balances) . However, this trend was altered during the first quarter of 2015. As of 3/31/2015, AGNC had a net long TBA MBS position of only $4.9 billion. Over the past several years, when AGNC lowered the company's net long TBA MBS position, management typically increased its on-balance regular MBS portfolio. This trend was apparent during the first quarter of 2015 as AGNC reported quarterly ERTI that was modestly higher when compared to amounts reported during 2014 when the company had a higher net long TBA MBS position. Since AGNC lowered the company's net long TBA MBS position during the first quarter of 2015, by taking these investments onto the balance sheet, this directly led to an increase in quarterly ERTI with an offsetting decrease in net dollar roll income.

However, beginning in the second quarter of 2015, AGNC had a more "cautious" stance regarding leverage due to MBS valuations and general macroeconomic events. As such, while AGNC's net long TBA MBS position only slightly increased to $6.9 billion as of 6/30/2015, the company's average on-balance sheet regular MBS portfolio decreased from $56.9 billion during the first quarter of 2015 to $53.7 billion during the second quarter of 2015 (based on par). When calculated, AGNC increased the company's net long TBA MBS position by $2.0 billion while decreasing its average on-balance sheet regular MBS portfolio by ($3.2) billion. As such, when compared to the prior quarter, AGNC's quarterly ERTI figure had a notable decrease for the second quarter of 2015.

This trend continued during the third quarter of 2015. While AGNC's net long TBA MBS position slightly increased to $7.1 billion as of 9/30/2015, the company's average on-balance sheet regular MBS portfolio decreased from $53.7 billion during the second quarter of 2015 to just $46.6 billion during the third quarter of 2015. When calculated, AGNC increased the company's net long TBA MBS position by only $0.2 billion while decreasing its average on-balance sheet regular MBS portfolio by ($7.1) billion. Therefore, AGNC's quarterly ERTI figure had another notable decrease for the third quarter of 2015. Put other way, a decrease in AGNC's average leverage during the quarter negatively impacted quarterly ERTI.

This trend modestly reversed course during the fourth quarter of 2015. AGNC's net long TBA MBS position once again slightly increased to $7.3 billion as of 12/31/2015. However, contrary to the trend that occurred during the prior two quarters, AGNC's average on-balance sheet regular MBS portfolio increased from $46.6 billion during the third quarter of 2015 to $49.9 billion during the fourth quarter of 2015. As such, AGNC increased both the company's net long TBA MBS position and its on-balance sheet regular MBS portfolio during the fourth quarter of 2015. This was the main reason why quarterly ERTI increased when compared to the prior quarter.

With that being said, AGNC's on-balance sheet regular MBS portfolio was still well below the historical average reported during 2012 - the first half of 2014. Since AGNC had quarterly common stock dividend distributions of ($204) million for the fourth quarter of 2015, this calculates to a quarterly overpayment of ($113) million and a payout ratio of 224%. Simply put, this was another material quarterly overpayment. However, as pointed out earlier, due to a specific scenario that has unfolded regarding AGNC's true earnings power and capital loss carryforward balance, the company's quarterly ERTI, on a standalone basis, is currently not the best metric when determining dividend sustainability. Due to the attractive yields and net dollar roll income being generated from AGNC's net long TBA MBS position, the company has stated its dividend will continue to be based on the earnings of the entire MBS portfolio (both on- and off-balance sheet MBS) for the foreseeable future.

Since AGNC's net long TBA MBS position is basically an extension of the balance sheet, I believe all net dollar roll income generated from these investments should be added to the company's quarterly ERTI figure to better visualize the earnings of the entire portfolio. Dollar roll income (loss) from a net long (short) TBA MBS position is treated as a reduction (addition) of an entity's cost basis per the IRC. As such, currently all net dollar roll income is being offset against AGNC's 2013 capital loss carryforward.

Still using Table 2 as a reference, AGNC reported quarterly ERTI and net dollar roll of $0.42 per common share for the fourth quarter of 2015. It should be noted this was the exact per common share amount I projected for the fourth quarter of 2015. When compared to the prior quarter, AGNC had a minor increase in the company's quarterly ERTI and net dollar roll figure. An explanation for this increase was provided when AGNC's quarterly ERTI figure was discussed several paragraphs earlier.

During the second, third, and fourth quarters of 2014, AGNC had a payout ratio which remained within a consistent range of 91% - 97%. However, this particular payout ratio increased to 104% for the first quarter of 2015. At the time, I believed this was still not an "alarming" ratio but did need to be monitored more closely in future quarters. During the second and third quarters of 2015, this ratio notably increased to 129% and 158%, respectively. For the fourth quarter of 2015, AGNC had ERTI and net dollar roll available to common shareholders of $144 million. As stated earlier, AGNC had common stock dividend distributions of ($204) million. When calculated, this was a quarterly overpayment of ($60) million and a payout ratio of 142%. As one can see, over time AGNC went from having a minor underpayment of quarterly ERTI and net dollar roll to a notable quarterly overpayment.

When looking at the results from TEST 1, I believe the probability of AGNC being able to maintain the company's monthly dividend of $0.20 per common share remained the same when compared to the prior quarter. One factor for this determination is AGNC's recent "leveling-off" of the company's "at risk" (total) leverage ratio. AGNC had an average at-risk (total) leverage ratio of 6.2x during the third quarter of 2015. This was the lowest average at-risk (total) leverage ratio since AGNC's initial public offering ("IPO") in 2008. AGNC's average at-risk (total) leverage ratio increased to 6.8x during the fourth quarter of 2015. While AGNC increased the company's average at-risk (total) leverage when compared to the prior quarter, this leverage ratio was still below its historical average. Due to events that have unfolded during the first quarter of 2016, management likely continued to increase AGNC's leverage.

With that being said, TEST 1 does not specifically account for AGNC's cumulative UTI/cumulative UTI and net dollar roll balance. I believe it would only be prudent to now perform TEST 2 and see if similar results can be obtained.

TEST 2 - Cumulative UTI/Cumulative UTI and Net Dollar Roll Coverage of Quarterly Dividend Distributions Ratio Analysis:

Before we begin TEST 2 of AGNC's dividend sustainability and projection analysis, let us first get accustomed to the information provided in Table 3 below. Table 3 is an extension of the information provided in Table 2 above. Table 3 below compares AGNC's quarterly dividend distributions to the company's cumulative UTI/cumulative UTI and net dollar roll balance for the same quarter. All figures within Table 3 are for the "three-months ended" (quarterly) timeframe except for AGNC's cumulative UTI/cumulative UTI and net dollar roll balance.

Table 3 - AGNC Cumulative UTI/Cumulative UTI and Net Dollar Roll Coverage of Quarterly Dividend Distributions Ratio Analysis (TEST 2)

Click to enlarge

(Source: Table created entirely by myself, partially using AGNC data obtained from the SEC's EDGAR Database [link provided below Table 2])

Cumulative UTI is a "running balance" that either increases or decreases each quarter as AGNC underpays (overpays) the company's dividend distributions. This type of direct relationship also occurs when analyzing AGNC's cumulative UTI and net dollar roll balance.

TEST 2 - Analysis and Results:

Using Table 3 above as a reference, I take AGNC's "cumulative UTI" figure (see red reference "M" in Table 3 above) and divide this amount by the quarterly "distributions to shareholders from ERTI" figure (see red reference "I" in Table 3 above). From this calculation, AGNC's "cumulative UTI coverage of quarterly dividend distributions ratio" is obtained (see red reference "N" in Table 3 above). The higher this ratio becomes, the more positive the results regarding AGNC's near-term dividend sustainability. Basically, this ratio shows the amount of cumulative UTI covering the current quarter's dividend distributions (after taking the current quarter's dividend distributions into account).

As of 9/30/2015, AGNC reported a cumulative "undistributed taxable deficit" ("UTD") balance of ($589) million. This deficit balance increased to ($702) million as of 12/31/2015. As such, there was a notable cumulative UTD increase during the fourth quarter of 2015. Due to the fact there were 9.0 million outstanding shares of common stock repurchased during the fourth quarter of 2015, AGNC needed to achieve a slightly lower quarterly ERTI "break-even" when compared to the prior quarter. When SOLELY looking at AGNC's cumulative UTD balance as of 12/31/2015, one could "easily" come to the conclusion the company needed to materially reduce dividend distributions as early as January 2016. However, since this cumulative UTD balance EXCLUDES the impact of AGNC's cumulative net dollar roll, I continue to believe this is an incorrect assumption.

Now let us take a look at TEST 2 when INCLUDING AGNC's cumulative net dollar roll. As of 12/31/2014, AGNC had a cumulative UTD and net dollar roll balance of $651 million. This balance reversed course during the first, second, third, and fourth quarters of 2015 and decreased to $455 million as of 12/31/2015.

So, what I believe should be interpreted from TEST 2 is that one should not SOLELY look to AGNC's cumulative UTD balance of ($702) million as of 12/31/2015 or the company's cumulative UTD and net dollar roll balance of $455 million as definitive evidence whether the "Board of Directors" ("BoD") will change the company's monthly dividend per share rate. As such, I currently believe TEST 1 has a heightened level of importance when compared to TEST 2. AGNC's near-term dividend sustainability will directly depend on the company's use of leverage, off-balance sheet net long (short) TBA MBS position, on-balance sheet regular MBS portfolio, and hedging coverage ratio.

With that being said, as has been consistent with the prior seven quarters, AGNC currently does not have a cumulative UTI "cushion" per se regarding quarterly overpayments. If AGNC's quarterly ERTI and net dollar roll cannot cover the company's quarterly dividend distributions, then I believe the BoD will adjust the monthly dividend per share rate accordingly. This notion was not an issue during 2014 as AGNC's quarterly ERTI and net dollar roll slightly exceeded quarterly dividend distributions. However, when management stated AGNC was taking more of a "defensive stance" regarding the use of leverage during 2015, the BoD made the cautious, yet prudent move to modestly lower the monthly dividend rate from $0.22 per share to $0.20 per common share beginning in May 2015 (as a "forward thinking" measure). Since AGNC reported quarterly ERTI and net dollar roll of $0.42 per common share for the fourth quarter of 2015, I believe AGNC's dividend continues to have heightened risk of being reduced in the future.

Brief Analysis/Discussion of MTGE's Dividend Sustainability (When Compared to AGNC):

Using Table 4 below as a reference, let us briefly compare AGNC to the company's affiliate MTGE regarding TEST 1 and TEST 2.

Table 4 - AGNC Versus MTGE Dividend Sustainability Analysis (TEST 1 and TEST 2)

Click to enlarge

(Source: Table created entirely by myself, partially using AGNC and MTGE data obtained from the SEC's EDGAR Database [link provided below Table 2])

Let us first discuss TEST 1. Using Table 4 above as a reference, when compared to AGNC, MTGE had a modestly higher quarterly ERTI per common share figure for the fourth quarter of 2015 (when excluding each company's net dollar roll amount). AGNC had quarterly ERTI of $0.27 per common share while MTGE had quarterly ERTI of $0.38 per common share. AGNC had a quarterly overpayment of ($113) million for the fourth quarter of 2015 while MTGE had a quarterly overpayment of only ($0.4) million. When calculated, AGNC had a quarterly payout ratio of 224% while MTGE had a quarterly payout ratio of 102%.

Let us now compare each company's results when net dollar roll activity is included. When compared to AGNC, MTGE had the same quarterly ERTI and net dollar roll per common share figure for the fourth quarter of 2015. AGNC and MTGE had quarterly ERTI and net dollar roll of $0.42 per common share. However, due to the fact AGNC had monthly dividends totaling $0.60 per common share for the fourth quarter of 2015 while MTGE had a quarterly dividend of $0.40 per common share, AGNC had a quarterly ERTI and net dollar roll overpayment of ($60) million while MTGE had a quarterly underpayment of $1.3 million. When calculated, AGNC had a quarterly payout ratio of 142% while MTGE had a quarterly payout ratio of only 93%. During 2015, AGNC reduced the company's monthly dividend rate from $0.22 per common share to $0.20 per common share while MTGE reduced its quarterly dividend rate from $0.65 per common share to $0.40 per common share. When calculated, AGNC reduced the company's quarterly dividend by ($0.06) per common share during 2015 while MTGE reduced its quarterly dividend by ($0.25) per common share. This was a notable difference. MTGE's material dividend reduction during the first and third quarters of 2015 allowed the company to be more "in-line" with reported quarterly ERTI and net dollar roll over the past four quarters. The same cannot be said about AGNC's quarterly ERTI and net dollar roll figure for past three quarters.

Now let us discuss TEST 2. Still using Table 4 above as a reference, when compared to AGNC, MTGE's cumulative UTI coverage of quarterly dividend distributions ratio was materially higher as of 12/31/2015 (excluding each company's net dollar roll activity). AGNC's cumulative UTI coverage of quarterly dividend distributions ratio was a factor of (3.44) while MTGE's ratio was a factor of 0.09 as of 12/31/2015.

Let us now compare results when each company's net dollar roll activity is included. When compared to AGNC, MTGE's cumulative UTI and net dollar roll coverage of quarterly dividend distributions ratio was modestly higher as of 12/31/2015. AGNC's cumulative UTI and net dollar roll coverage of quarterly dividend distributions ratio was a factor of 2.33 as of 12/31/2015 while MTGE's ratio was a factor of 3.01. It should be noted AGNC's prior outperformance regarding this specific metric "reversed course" during the third quarter of 2015.

From the analysis provided above, including additional factors not discussed within this article, I project MTGE will declare a stable $0.40 per common share dividend for the first quarter of 2016.

Brief Analysis/Discussion of NLY's Dividend Sustainability:

NLY reported quarterly ERTI of $0.31 per common share and quarterly ERTI and net dollar roll of $0.41 per common share for the second quarter of 2015. NLY's quarterly ERTI and net dollar roll is also known as the company's "estimated core earnings" ("ECE"). As such, the probability of NLY being able to maintain the company's dividend of $0.30 per common share seemed fairly logical.

However, NLY reported quarterly ERTI of only $0.11 per common share and quarterly ECE of only $0.21 per common share for the third quarter of 2015. With that being said, it should also be noted management provided an additional metric in regards to NLY's dividend during the third quarter of 2015. This additional metric was "normalized core earnings" ("NCE"). When including the impacts from NLY's "catch-up" premium amortization expense, the company reported NCR available to common shareholders of $0.30 per share during the third quarter of 2015. This per common share amount matched NLY's dividend distributions during the quarter. Even though NLY implied this metric would factor into the company's dividend per share rate for the fourth quarter of 2015, I would caution readers that a company's premium amortization expense is, in fact, a "non-cash" item. Per the accrual method of accounting, which is based on "Generally Accepted Accounting Principles" ("GAAP"), MBS investments are valued at "par" at the time of acquisition and all related premiums (discounts) are amortized over the estimated life of the securities. Per the tax method of accounting, which is based on the "Internal Revenue Code" ("IRC"), MBS investments are valued at the actual purchase price and the investment's "cost basis" remains unchanged over the life of the securities in most instances (not per a monetary standpoint due to principle repayments but as a percentage to par). Simply put, timing differences between the two sets of accounting methodologies occur. Therefore, NLY's NCE metric is more of a reconciling GAAP line item since this directly impacts a company's premium amortization expense (discount accretion income) account over the estimated life of the MBS portfolio.

With that being said, NLY's "Board of Directors" ("BoD") ultimately decides on the quarterly dividend per share rate. NLY's BoD may continue to utilize NCE as a dividend per share rate metric. Along with several other recent trends, the notion of NLY's BoD considering the company's quarterly NCE was one of the main reasons why I correctly projected a stable dividend of $0.30 per common share for the fourth quarter of 2015. Even though I believe it is a low probability that NCE will be higher than ECE during the fourth quarter of 2015, I believe a discussion of this topic is beneficial for readers.

Through a detailed quarterly ERTI, ECE, and NCE projection model, I am projecting NLY will declare a stable $0.30 per common share dividend for the first quarter of 2016.

During the prior quarter, I believed NLY declaring a stable dividend for the fourth quarter of 2015 was more of a modest (65%) probability. However, I believe NLY declaring a stable dividend for the first quarter of 2016 has increased to a high (75%) probability. I believe NLY declaring a dividend of $0.25 per common share has a low (25%) probability of occurring.

Conclusions Drawn:

To sum up all the information in this article, TEST 1 showed AGNC had ERTI available to common shareholders of $91 million for the fourth quarter of 2015. When including activities within AGNC's net long TBA MBS position, the company had ERTI and net dollar roll of $144 million. Since AGNC had common stock dividend distributions of ($204) million during the fourth quarter of 2015, this calculated to an overpayment of ($60) million and a payout ratio of 142%. While this was an improvement when compared to an overpayment of ($76) million and a payout of 158% for the third quarter of 2015, AGNC's overpayment for the fourth quarter of 2015 should still be seen as a negative trend for the company's future dividend sustainability.

TEST 2 showed AGNC had a cumulative UTI coverage of quarterly dividend distributions ratio of (3.44) as of 12/31/2015. When solely looking at this metric, one would expect an immediate material reduction to AGNC's dividend. However, when including the net dollar roll being generated from AGNC's net long TBA MBS position, the company had a cumulative UTI and net dollar roll coverage of quarterly dividend distributions ratio of 2.23 as of 12/31/2015. So, what I believe should be interpreted from TEST 2 is that one should not SOLELY look to AGNC's cumulative UTD balance of ($702) million as of 12/31/2015 or the company's cumulative UTI and net dollar roll balance of $455 million as definitive evidence whether the BoD will change the company's monthly dividend per share rate. As such, I currently believe TEST 1 has a heightened level of importance when compared to TEST 2.

AGNC's management team has continued to state the company's dividend will be based on the earnings of the entire MBS portfolio (both on- and off-balance sheet portfolios) for the foreseeable future. As such, I believe AGNC's near-term dividend sustainability will directly depend on (but not limited to) the company's use of leverage, off-balance sheet net long (short) TBA MBS position, on-balance sheet regular MBS portfolio, and hedging coverage ratio.

Since AGNC reported quarterly ERTI and net dollar roll of $0.42 per common share for the fourth quarter of 2015, I believe AGNC's dividend continues to have heightened risk of being reduced in the future. However, even though most signs point to a dividend reduction, AGNC's BoD has a tough decision to make in my opinion.

If the BoD makes the "prudent move" and lowers the dividend, I believe it is a high likelihood that AGNC's stock price will be negatively impacted over the short term. I believe it's a foregone conclusion that management/the BoD is well aware of what would likely occur to AGNC's stock price if the company announces another dividend reduction. In addition, management/the BoD also needs to consider what sector peers are declaring to remain as an attractive equity investment to market participants. I believe one sector peer AGNC is usually compared to is NLY. This mREIT has continued to maintain a stable dividend over the past two years. Simply put, I believe NLY's stable dividend has put pressure on AGNC to also maintain the company's dividend per share rate.

With that being said, the following are my "best case", "worst case", and most probable AGNC common stock dividend projection for February 2016 - March 2016 (including corresponding probabilities):

AGNC's Best Case Scenario for February 2016 - March 2016: Dividend of $0.20 per common share (Probability of 60%)

AGNC's Worst-Case Scenario for February 2016 - March 2016: Dividend of $0.16 per common share (Probability of 10%)

AGNC's "Middle-of-the-Road" Scenario for February 2016 - March 2016: Dividend of $0.17 - $0.19 per common share (Probability of 30%)

This projection is based on the assumption of AGNC's "reversal" of prior trends by continuing to increase leverage when compared to levels witnessed during 2015. I am projecting AGNC to increase leverage due to the decreased likelihood of the Federal Open Market Committee ("FOMC") increasing the Federal ("Fed") Funds Rate during 2016. Higher leverage would positively impact AGNC's quarterly ERTI and net dollar roll.

My BUY, SELL, or HOLD Recommendation:

From the analysis provided above, including additional factors not discussed within this article, I currently rate AGNC and NLY as a SELL when I believe the company's stock price is trading at less than a (15%) discount to CURRENT BV (BV as of 2/5/2016), a HOLD when trading at or greater than a (15%) but less than a (25%) discount to CURRENT BV, and a BUY when trading at or greater than a (25%) discount to CURRENT BV.

As such, I currently rate AGNC and NLY as a HOLD since each company's stock price is trading at or greater than a (15%) but less than a (25%) discount to my projected CURRENT BV.

I currently rate MTGE as a SELL when I believe the company's stock price is trading at less than a (20%) discount to CURRENT BV, a HOLD when trading at or greater than a (20%) but less than a (30%) discount to CURRENT BV, and a BUY when trading at or greater than a (30%) discount to CURRENT BV. As such, I currently rate MTGE as a BUY since the stock is trading at or greater than a (30%) discount to my projected CURRENT BV.

This recommendation considers the following mREIT factors: 1) projected future MBS price movements; 2) projected future derivative valuations; and 3) projected near-term dividend per share rates. This recommendation also considers the recent lowered probability of several Fed Funds Rate increases by the FOMC during 2016 due to recent macroeconomic trends/events.

Final Note: Each investor's BUY, SELL, or HOLD decision is based on one's risk tolerance, time horizon, and dividend income goals. My personal recommendation will not fit each reader's current investing strategy. Within the past 90 days, I have not directly increased or decreased my AGNC position (only through reinvested dividends). On 11/27/2015, I initiated a position in AGNC's "Series B Preferred Stock" which trades under the symbol "AGNCB". Subsequent to 11/27/2015, I have selectively increased my position in AGNCB. When combined, my current AGNCB position has a weighted average price of $23.215 per share. I currently hold 0.70% of the outstanding shares of AGNC's Series B Preferred Stock. Each trade was disclosed to readers in real time (that day) via the "StockTalks" feature of Seeking Alpha. This weighted average per share price excludes all dividends received/reinvested.

Disclosure: I am/we are long AGNC, MTGE.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: I currently have no position in NLY.