Seeking Alpha
Long-term horizon, dividend investing, mREITs, BDCs
Profile| Send Message| ()  

Focus of Article:

The focus of this article is to provide a mid-fourth quarter update on American Capital Agency Corp. (AGNC) regarding its mortgage-backed security ('MBS') and derivative portfolios. Each portfolio will be separately analyzed according to its composition and valuation. I feel this mid-quarter update will provide readers a general direction on how the first-half of the fourth quarter of 2013 has panned out regarding AGNC's MBS and derivative strategies. This article will mainly compare and contrast what has occurred so far in the fourth quarter of 2013 versus the third quarter of 2013 regarding AGNC's MBS and derivative portfolios. This will also help readers understand how the second-half of the current quarter could pan out as interest rates fluctuate.

I would like to first analyze AGNC's MBS portfolio as of 9/30/2013 and identify the changes AGNC made in the prior quarter which will impact the current quarter. In conjunction with this portfolio, specific quarterly MBS prices will be analyzed and discussed. This information will assist in projecting AGNC's MBS portfolio valuation as of 11/15/2013. After its MBS portfolio is analyzed and discussed, I will then analyze AGNC's derivative portfolio as of 9/30/2013 and identify the changes the company made in the prior quarter which will impact the current quarter. In conjunction with this portfolio, specific quarterly MBS prices, swap rates, and U.S. Treasury yields will be analyzed and discussed. This information will assist in projecting a valuation for AGNC's derivative portfolio as of 11/15/2013.

Due to the length of the material covered in this article, I feel it is necessary to break this particular article into three parts. This article will be broken-down by the following two account portfolios:

PART 1:

- AGNC's MBS Portfolio (Composition and Valuation Analysis)

PART 2:

- AGNC's Derivative Portfolio (Composition Analysis)

PART 3:

- AGNC's Derivative Portfolio (Valuation Analysis)

- Book Value ('BV') Projection (BV as of 11/15/2013)

Author's Note: This three-part article is a very detailed look at AGNC's MBS and derivative portfolios. I perform this detailed analysis for readers who anticipate/want such an analysis performed each quarter. For readers who just want the summarized conclusions/results, I would suggest to just scroll down to the "Conclusions Drawn" section at the bottom of the each part of the article.

Side Note Regarding Using Prior Quarter's Ending MBS Balances as a Current Quarter Valuation Technique:

Prior to showing AGNC's 9/30/2013 MBS portfolio via Table 1 (see below), I feel a discussion regarding why I use this specific valuation technique should be addressed.

A few readers in the comments section of a few of my past mREIT articles (including other authors' articles) highlighted that an mREIT will have numerous quarterly activities which will change a MBS portfolio within any stated quarter. Therefore, these specific readers feel it is basically useless to even try and value an mREIT's MBS portfolio (or any account for that matter) in a current quarter. I would have to firmly disagree with this notion.

The asset valuations of any account on the balance sheet that represent the prior quarter's ending balance are definitive in nature. Regarding an mREIT's MBS portfolio, even if some of these securities are sold in the current quarter, they still must be accounted for. They simply are not "washed away" and disposed of without having an accounting treatment. Regarding this particular topic, all of AGNC's MBS have quarterly fair market value ('FMV') adjustments (also known as "mark-to-market" adjustments) which will either be "realized" or "unrealized" by the end of the fourth quarter of 2013 (FMV as of 12/31/2013). As such, knowing the FMV of these MBS as of 9/30/2013 is fairly important.

Therefore, the only aspects left open to interpretation are the amount of MBS sales and the amount of MBS purchases that occur in a given quarter (including at what coupon rates and maturities). This is where a level of "projection" based on certain "assumptions" must be taken into consideration. AGNC's entire quarterly MBS valuation change will be accounted for in either of the following two income statement accounts: 1) gain (loss) on sale of agency securities, net or 2) unrealized gain (loss) on available-for-sale securities, net. The balance sheet account that would be affected is the "agency securities, at fair value" account. Through detailed research and data compilation, one can project (to a reasonable degree) how management "should" act within any given quarter regarding purchases and sales. However, I stress beforehand this will not be an "exact science" each quarter. There will be some variances that occur in a quarter if more/less sales and/or purchases actually occur than originally projected. Additionally, unanticipated quarterly changes in the percentage of coupons/maturities held within the MBS portfolio would cause a slight deviation in asset valuations.

Regarding the company's most recent quarter, AGNC aggressively sold a proportionally higher amount of its lowest coupon holdings to further offset future valuation losses on its MBS portfolio in a rising interest rate environment. Such an aggressive MBS strategy early in the quarter was a deviation from past quarter activities. Furthermore, management performed a majority of its MBS sales when interest rates were near the highest point during the quarter thus increasing the amount of its realized losses. I personally felt this was poor timing on behalf of management.

When combining AGNC's gain (loss) on sale of agency securities, net and unrealized gain (loss) on available-for-sale securities, net accounts together, I projected a valuation gain of $330 million for the third quarter of 2013. My projected range for these two combined accounts were ($220) - $880 million. AGNC reported a combined realized and unrealized valuation gain of $100 million for the third quarter of 2013. This calculates to an over-projection of $230 million regarding AGNC's realized and unrealized MBS valuation accounts. As such, actual results were still within my stated range but more towards the lower end of my range. Therefore, several of my assumptions proved to be incorrect regarding the third quarter of 2013. Again, it was not the valuation technique itself that was flawed but several of the assumptions I used.

The main assumption that was incorrect was management performing MBS sales when market interest rates were towards the lower end of the range during the quarter. As stated earlier, management performed a majority of its MBS sales when interest rates were near the highest point during the quarter thus increasing the amount of its realized losses. I personally felt this was poor timing on behalf of management.

Backing a step back, all income statement account valuation figures I projected were within my stated ranges for the quarter. However, most reported figures trended near the lower end of my ranges. As such, overall cumulative results were lower than I expected.

With that being said, one can still "reasonably" predict how management "should" perform regarding such activities in a specified quarter. This should be evidenced by my past articles regarding AGNC's MBS valuation projections for the second quarter of 2013. On a valuation basis, when including both realized and unrealized valuation losses regarding AGNC's MBS portfolio, I was basically "spot-on" with my projected ($2.8) billion in MBS valuation losses. AGNC's management took the appropriate steps in the second quarter of 2013 to try and minimize BV losses given its MBS portfolio as of 3/31/2013. The following are links to my second quarter of 2013 income statement and BV articles where such MBS valuations and assumptions were accurately projected:

American Capital Agency Corp.'s Upcoming Q2 2013 Income Statement Projection (Part 1)

American Capital Agency Corp.'s Upcoming Q2 2013 Income Statement Projection (Part 2)

American Capital Agency Corp.'s Upcoming Q2 2013 Book Value Projection (As of June 30, 2013)

To appropriately begin a current quarter's account valuation analysis, this specific technique has proved to be the best methodology in my years of research and data compilation. If a similar valuation analysis was attempted without using this specific methodology, material valuation variances would most likely occur in each instance.

A) AGNC's MBS Portfolio - Composition and Valuation Analysis:

Let us first understand AGNC's MBS portfolio as of 9/30/2013 regarding its composition. This will include a net change analysis when comparing its 9/30/2013 balance to its 6/30/2013 balance. This will be followed by an MBS portfolio valuation analysis for the first-half of the fourth quarter of 2013.

Side Note: AGNC has continued to have an overwhelming portion of its MBS portfolio in fixed-rate agency holdings (approximately 97% of its MBS portfolio as of 9/30/2013). As such, all references to AGNC's MBS portfolio below will automatically be implied as being both "fixed-rate" and "agency" MBS unless otherwise noted.

1) AGNC's MBS Portfolio - Composition Analysis (As of 9/30/2013; Including Net Change Analysis):

Table 1 below shows AGNC's MBS portfolio as of 9/30/2013. All MBS are sorted by its maturity (15, 20, or 30-year) and coupon rate (2.5% - 6%). Table 1 displays the following columns (in order from left to right): 1) MBS coupon rate; 2) par value; 3) amortized cost value (basis); 4) market value ('FMV'); 5) percentage of Lower Loan Balance ('LLB') or Home Affordable Refinance Program ('HARP') MBS; 6) amortized cost basis as a percentage to par; 7) FMV as a percentage to par; 8) weighted average coupon ('WAC') rate; 9) weighted average yield; 10) average age of MBS; 11) monthly constant prepayment rate (CPR); and finally 12) weighted average duration of MBS.

Table 1 - AGNC MBS Portfolio (As of 9/30/2013)

(click to enlarge)

Having Table 1 above as a reference, let us first discuss how AGNC's MBS portfolio as of 9/30/2013 compared to its MBS portfolio as of 6/30/2013 regarding its overall composition. This will enable us to better understand what changes were made during the third quarter of 2013. This will ultimately help us when performing a mid-fourth quarter of 2013 valuation analysis on AGNC's MBS portfolio as of 9/30/2013.

To better understand how AGNC's MBS portfolio as of 9/30/2013 changed when compared to its MBS portfolio as of 6/30/2013, Table 2 is shown below. Table 2 provides key information on some recent quarterly net changes to AGNC's MBS portfolio regarding the following: 1) par value; 2) FMV versus amortized cost valuation; 3) amortized cost basis percentage; 4) FMV MBS price percentage; 5) WAC percentage and 6) CPR percentage. Table 2 also breaks out these quarterly net changes into their respective coupon rates. Due to the immateriality of AGNC's 20-year MBS, only the company's 15 and 30-year holdings will be discussed below.

Table 2 - AGNC MBS Portfolio Quarterly Net Changes (9/30/2013 vs. 6/30/2013)

(click to enlarge)

Using Table 2 above as a reference, AGNC sold a rather large portion of its 30-year 3.0% and 3.5% coupon MBS during the third quarter of 2013. The net quarterly change within AGNC's 30-year 3.0% and 3.5% coupon MBS was ($10.3) and ($5.5) billion, respectively. AGNC stated this was due to its aggressive portfolio realignment into less price sensitive 15-year MBS. Rather than sustain additional valuation losses within these specific coupons in a rising interest rate environment, AGNC decided to sell a large portion of its 30-year 3.0% and 3.5% coupon MBS. In the current environment, the lowest coupon MBS have been (and will continue to be) the most susceptible to price fluctuations as interest rates move in either direction. This would help AGNC offset future additional valuation losses if MBS prices were further suppressed in the coming quarters due to rising interest rates.

AGNC also sold a material portion of its 15-year 2.5% and 3.0% coupon MBS during the third quarter of 2013. The net quarterly change within AGNC's 15-year 2.5% and 3.0% coupon MBS was ($3.8) and ($2.8) billion, respectively. Again, AGNC is trying to mitigate future valuation losses if interest rates rise in current and future quarters. Again, as the coupon rate of a particular existing MBS drifts lower and farther away from the current coupon rate offered on a similar type of MBS in the market, the existing lower coupon MBS escalates its valuation losses. Instead of incurring continued unrealized losses in current and future quarters, AGNC continued to quickly sell these 15-year 2.5% and 3.0% coupon MBS for a material realized loss in the third quarter of 2013. It should also be noted a material portion of the recently sold lower coupon MBS was newly converted TBA MBS holdings. Since AGNC's TBA MBS portfolio is accounted for within its derivatives account, these holdings will be discussed during PART 2 and PART 3 of this article.

Therefore, in selling these types of lower coupon MBS, AGNC offset future valuation losses and also has new capital to deploy into the currently higher-yielding, less price sensitive 15-year MBS with a coupon rate of 3.5% (purchases will discussed shortly). AGNC could also wait regarding deploying this newly obtained capital if management feels rates will continue to rise rather quickly (wait till a higher coupon MBS becomes available and its price is attractive). In the meantime, AGNC can have a net long U.S. Treasury securities position instead of keeping unused capital in extremely low interest bearing cash accounts. When looking at AGNC's balance sheet over the past three quarters, this is exactly what management had done. As of 3/31/2013, AGNC did not have a position within its "U.S. Treasuries, at fair value" asset account. However, as of 6/30/2013, AGNC had a FMV of $3.7 billion within this account. As of 9/30/2013, this account balance grew to a FMV of $4.8 billion. It appears management is being cautious as to what assets or under what conditions they would fully deploy unused capital.

Regarding purchases, Table 2 reveals that AGNC made some material acquisitions of 15-year 3.5% coupon MBS. The determination of which MBS coupons were most readily available in a given quarter is in direct correlation to where recent fixed mortgage rates were. The net quarterly change within AGNC's 15-year 3.5% coupon MBS was an increase of $8.2 billion. It seems all other MBS coupon rates had immaterial net price changes and will not be mentioned within this article.

Still using Table 2 above as a reference, when comparing its quarterly net changes, AGNC had a net increase of $1.0 billion regarding its 15-year MBS holdings. Also, AGNC had a net decrease of ($18.0) billion regarding its 30-year MBS holdings. Management has taken a "defensive posture" regarding the company's current MBS portfolio. This was a direct result of the volatile nature of the markets from rising interest rates during the latter half of the second quarter of 2013 continuing into the first half of the third quarter of 2013 regarding MBS price movements. AGNC made the conscious decision of lowering its exposure to 30-year MBS holdings which have more pronounced price movements when compared to its 15-year MBS counterparts with similar coupons. This was the main reason why AGNC continued to decide to shift a portion of its MBS holdings into 15-year MBS. AGNC is trying to "defend" its BV through the acquisition of less "price sensitive" 15-year MBS. From a BV perspective, this strategy makes sense and should be deemed a wise move if interest rates continue to rise.

As briefly stated above, it should be noted a majority of the decrease in AGNC's MBS portfolio was its shift was a net long TBA MBS position of $14.4 billion as of 6/30/2013 to a net short TBA MBS position of ($7.3) billion as of 9/30/2013. Since AGNC's TBA MBS portfolio is accounted for within its derivatives account, these holdings will be discussed during PART 2 and PART 3 of this article.

Now that we have an understanding of AGNC's MBS portfolio regarding its recent past and current composition, let us shift topics and now focus on AGNC's MBS portfolio regarding its valuation through the first-half of the fourth quarter of 2013.

2) AGNC's MBS Portfolio - Valuation Analysis (Through the First-Half of the Fourth Quarter of 2013):

Prior to performing a valuation analysis on AGNC's current MBS portfolio, let us first analyze the MBS price movements during first-half of the fourth quarter of 2013. This analysis will first look at 15-year fixed-rate agency MBS price movements. This will then be followed by a similar analysis of 30-year fixed-rate agency MBS price movements. By first analyzing the 15 and 30-year MBS price movements, it will help us better understand how I come up with my projected AGNC MBS portfolio valuation figures discussed later in the article.

Table 3 - 15-Year Fixed-Rate Agency MBS Price Movements (Through the First-Half of the Fourth Quarter of 2013)

(click to enlarge)

Table 3 above shows the 15-year fixed-rate agency MBS price movements during the first-half of the fourth quarter of 2013. It breaks out the 15-year fixed-rate agency MBS holdings by government sponsored entity ('GSE'). This includes both Fannie Mae (OTCQB:FMCC) and Freddie Mac (OTCQB:FNMA) MBS. As of 9/30/2013, AGNC's Ginnie Mae holdings only accounted for 0.29% of the company's MBS portfolio. As such, Ginnie Mae fixed-rate agency MBS price movements are immaterial for projection purposes and thus excluded from this table and article. Table 3 further breaks out the 15-year fixed-rate agency MBS price movements into the various coupons on AGNC's books ranging from 2.5% - 4.5%. AGNC holds an immaterial balance of 15-year fixed-rate agency MBS over the 4.5% coupon and thus these specific coupons are excluded from Table 3 above.

From the information portrayed via Table 3 above, a valuation gain (loss) can be calculated which is broken down by the various coupons. This valuation gain (loss) is performed via Table 5 later in the article. As stated in a side note earlier in the article, an exact valuation figure cannot be obtained because AGNC continually changes its portfolio holdings in a given quarter. As such, I must determine specific purchase and sale assumptions towards the end of my valuation analysis.

Using Table 3 above as a reference, let us look at the 15-year fixed-rate agency MBS price movements for the current quarter regarding a few of the coupon rates in which AGNC holds its larger balances. The cumulative quarterly net MBS price movements for each coupon rate are shown within Table 3 under the "Cumulative Quarterly Change" column.

To illustrate, through the week ending on 11/15/2013, the Fannie 15-year fixed-rate agency MBS with a 2.5% coupon had a cumulative quarterly price decline of (0.25) to settle its price at 100.36 (100 being par). The Fannie 15-year fixed-rate agency MBS with a 3.0% coupon had a cumulative quarterly MBS price decline of (0.08) to settle its price at 103.45 (100 being par). The Fannie 15-year fixed-rate agency MBS with a 3.5% coupon (AGNC's largest 15-year fixed-rate agency MBS position as of 9/30/2013) had a cumulative quarterly MBS price increase of 0.01 to settle its price at 105.59 (100 being par).

Still using Table 3 above as a reference, the Fannie 15-year fixed-rate agency MBS with a coupon of 4.0% and 4.5% had slight to modest cumulative quarterly MBS price increases through the week ending on 11/15/2013. As such, when evaluating all coupons, it looks like AGNC's Fannie 15-year fixed-rate agency MBS had minimal cumulative quarterly MBS price changes for the first-half of the fourth quarter of 2013. Through the week ending on 11/15/2013, slight 15-year fixed-rate agency MBS price decreases occurred on the 2.5% and 3.0% coupons while slight MBS price increases occurred on the 3.5%, 4.0%, and 4.5% coupons.

For the Freddie 15-year fixed-rate MBS, the same conclusions hold true. When compared to the Fannie 15-year fixed-rate agency MBS, the Freddie 15-year fixed-rate agency MBS usually have slight differences in weekly (hence cumulative quarterly) price valuations across the same coupons. However, the same general themes typically hold true.

Now that we have an understanding of the 15-year fixed-rate MBS price movements through the first-half of the fourth quarter of 2013, let us take a look at the 30-year fixed-rate agency MBS price movements.

Table 4 - 30-Year Fixed-Rate Agency MBS Price Movements (Through the First-Half of the Fourth Quarter of 2013)

(click to enlarge)

Table 4 above shows AGNC's 30-year fixed-rate agency MBS price movements during the first-half of the fourth quarter of 2013. It breaks out the 30-year fixed-rate agency MBS holdings by GSE. As was the case with the 15-year fixed-rate agency MBS, this includes both Fannie Mae and Freddie Mac holdings. As stated earlier, AGNC's Ginnie Mae fixed-rate agency MBS holdings are immaterial for projection purposes. As such, Ginnie Mae fixed-rate agency MBS price movements are excluded from this table. Table 4 further breaks out the 30-year fixed-rate agency MBS price movements into the various coupons on AGNC's books ranging from 3.0% to 5.0%. AGNC holds an immaterial balance of 30-year fixed-rate agency MBS over the 5.0% coupon and thus these specific coupons are excluded from Table 4 above.

From the information portrayed via Table 4 above, a valuation gain (loss) can be calculated which is broken down by the various coupons. This valuation gain (loss) is performed via Table 5 later in the article. As stated in a side note earlier in the article, an exact valuation figure cannot be obtained because AGNC continually changes its portfolio holdings in a given quarter. As such, I must determine specific purchase and sale assumptions towards the end of my valuation analysis.

Using Table 4 above as a reference, let us look at the 30-year fixed-rate agency MBS price movements for the current quarter regarding a few of the coupon rates in which AGNC holds its larger balances. The cumulative quarterly net MBS price movements for each coupon rate are shown within Table 4 under the "Cumulative Quarterly Change" column.

To illustrate, through the week ending on 11/15/2013, the Fannie 30-year fixed-rate agency MBS with a 3.0% coupon had a cumulative quarterly price decline of (0.48) to settle its price at 97.22 (100 being par). The Fannie 30-year fixed-rate agency MBS with a 3.5% coupon (AGNC's largest 30-year fixed-rate agency MBS position as of 9/30/2013) had a cumulative quarterly MBS price decline of (0.34) to settle its price at 101.48 (100 being par). The Fannie 30-year fixed-rate agency MBS with a 4.0% coupon had a cumulative quarterly MBS price decline of (0.07) to settle its price at 104.78 (100 being par).

Still using Table 4 above as a reference, the Fannie 30-year fixed-rate agency MBS with a coupon of 4.5% and 5.0% had slight to modest cumulative quarterly MBS price increases through the week ending on 11/15/2013. As such, when evaluating all coupons, it looks like AGNC's Fannie 30-year fixed-rate agency MBS had minimal cumulative quarterly MBS price changes for the first-half of the fourth quarter of 2013. Through the week ending on 11/15/2013, slight 30-year fixed-rate agency MBS price decreases occurred on the 3.0%, 3.5%, and 4.0% coupons while slight MBS price increases occurred on the 4.5% and 5.0% coupons.

For the Freddie 30-year fixed-rate agency MBS, the same conclusions hold true. When compared to the Fannie 30-year fixed-rate agency MBS, the Freddie 30-year fixed-rate agency MBS usually have slight differences in weekly (hence cumulative quarterly) price valuations across the same coupons. However, the same general themes typically hold true.

Now that we have an understanding of the 15 and 30-year fixed-rate agency MBS price movements through the first-half of the fourth quarter of 2013, let us take a look at how these fixed-rate agency MBS price movements affect AGNC's MBS portfolio regarding the company's weekly and cumulative quarterly valuations through the week ending 11/15/2013 (quantified in dollar amounts).

Table 5 - AGNC Summarized Weekly and Cumulative Quarterly Valuation Gain (Loss) on its MBS Portfolio (Through the Week Ending 11/15/2013)

(click to enlarge)

Table 5 above shows AGNC's weekly and cumulative quarterly projected valuation gain (loss) on the company's 15 and 30-year Fannie and Freddie fixed-rate agency MBS holdings across all coupons (including immaterial coupons omitted from Table 3 and Table 4 above). For this specific analysis, Table 5 shows the projected valuation gain (loss) through the week ending 11/15/2013 and quantifies the cumulative quarterly gain (loss) in dollar amounts.

Side Note: As stated earlier in the article, the company's Ginnie Mae holdings only accounted for 0.29% or $249 million of AGNC's MBS portfolio as of 9/30/2013. Also, AGNC's 20-year fixed-rate agency MBS holdings only accounted for 1.80% or $1.4 billion of AGNC's MBS portfolio as of 9/30/2013. For purposes of this analysis, both these amounts are deemed immaterial and excluded from Table 5 above. However, when I compute my quarterly income statement and BV projections for AGNC, these immaterial holdings are considered and included in all valuation calculations (even if not directly shown within a specified table).

Specifically looking through the week ending 10/11/2013, AGNC had a projected cumulative quarterly valuation loss of ($426) million on its MBS portfolio. The second week of October had mortgage rates/U.S. Treasury yields rise spurred by the looming U.S. Government shutdown and possible U.S. default on its credit obligations. As such, markets began to anticipate and "price in" a quick rise in rates. However, during the week ending 10/18/2013, mortgage rates/U.S. Treasury yields sharply reversed course once the U.S. debt ceiling crisis was averted and the U.S. Government shutdown ended. As such, AGNC had a projected cumulative quarterly valuation gain of $103 million on its MBS portfolio through the week ending 10/18/2013.

Through the week ending 11/15/2013, I am projecting AGNC had a cumulative quarterly valuation loss of approximately ($70) million. This was due in part to two pieces of data over the past few weeks. First, due to a much stronger than expected October Employment Situation Report, mortgage rates/U.S. Treasury yields "spiked" during the week of 11/8/2013. Due to the "better-than-expected" results from this economic indicator, markets reacted by once again considering the notion of the Federal Reserve ('Fed') beginning to end its Quantitative Easing Bond Purchasing Program (QE3) as early as December 2013. However, mortgage rates/U.S. Treasury yields somewhat reversed course during the week ending 11/15/2013 due to the "dovish" remarks by the incoming Fed Chairwoman Janet Yellen at the Senate confirmation hearings. As such, the markets reacted to the incoming Fed Chairwoman's comments by assuming a December 2013 "taper" of the Fed's QE3 program will not occur until 2014.

Since one can accurately track existing fixed-rate agency MBS price movements throughout the various coupons, a general sense of AGNC's cumulative quarterly valuation gain (loss) on the company's MBS portfolio can be achieved. As of 11/15/2013, when considering the impact of AGNC's specified pool holdings and quarterly MBS purchases, I have projected a valuation loss of approximately ($90) million regarding AGNC's MBS portfolio (prior to the reclassification of MBS sold during the quarter). Due to the volatile nature of short-term mortgage rates/U.S. Treasury yields, I stress readers need to closely watch future events as they unfold and understand the impacts of interest rate movements regarding the valuation of AGNC's MBS portfolio.

Brief Discussion of MTGE's MBS Portfolio (In Comparison to AGNC's MBS Portfolio):

AGNC's sister company, American Capital Mortgage Investment Corp. (MTGE), is also classified as a mortgage real estate investment trust (mREIT) which currently earns a majority of its income from investing (through leverage) in agency MBS. Approximately 88% of MTGE's MBS portfolio as of 9/30/2013 is in agency holdings (mostly fixed-rate MBS). MTGE further diversifies the company's investments with non-agency holdings including prime and subprime mortgage loans, option adjustable rate mortgages ('ARM'), and Alt-A loans (per management's discretion). Non-agency investments include residential mortgage-backed securities ('RMBS') backed by residential mortgages which are not guaranteed by a GSE or U.S. government agency. MTGE did not materially change the balance of its non-agency portfolio during the third quarter of 2013. MTGE had a total non-agency portfolio of $931 million as of 6/30/2013. As of 9/30/2013, this balance only declined ($3) million to $928 million. When compared to the company's agency MBS portfolio of $6.8 billion as of 9/30/2013, MTGE slightly increased its proportional share of its non-agency portfolio by 1% during the third quarter of 2013. As such, MTGE's proportional share on its non-agency portfolio increased from 11% of its total investment balance as of 6/30/2013 to 12% as of 9/30/2013.

MTGE's 9/30/2013 agency MBS portfolio mainly consisted of 15 and 30-year fixed-rate MBS that had similar characteristics when compared to AGNC's fixed-rate agency MBS portfolio. This includes a somewhat similar proportion of LLB and HARP loans and similar MBS coupon rates (some slight percentage differences).

One slight difference between AGNC's and MTGE's fixed-rate agency MBS portfolio that should be noted is the composition of 15 and 30-year fixed-rate agency holdings as of 9/30/2013 (as a percentage of its total fixed-rate agency MBS portfolio).

Table 6 - AGNC vs. MTGE Fixed-Rate Agency MBS Portfolio (As of 9/30/2013)

(click to enlarge)

Using Table 6 above as an indirect reference, AGNC increased its 15-year fixed-rate agency MBS holdings from 43% as of 6/30/2013 to 53% as of 9/30/2013 (when including AGNC's net long (short) TBA MBS position). AGNC decreased its 30-year fixed-rate agency MBS holdings from 57% as of 6/30/2013 to 45% as of 9/30/2013 (when including AGNC's net long (short) TBA MBS position). All remaining percentages were in 20-year fixed-rate MBS.

In contrast, MTGE increased its 15-year fixed-rate agency MBS holdings from 29% as of 6/30/2013 to 40% as of 9/30/2013. MTGE decreased its 30-year fixed-rate agency MBS holdings from 69% as of 6/30/2013 to 57% as of 9/30/2013. All remaining percentages were in 20-year fixed-rate MBS. As such, a slight difference in MBS valuations will occur between the companies' fixed-rate MBS portfolios. MTGE's fixed rate agency MBS portfolio has a slightly higher risk regarding possible BV losses if rates were to quickly rise. This is strictly from a fixed-rate agency MBS portfolio perspective and does not consider each company's offsetting hedges yet (which will be discussed in PART 2 and PART 3 of this article). I would either anticipate a slightly higher notional balance of swaps and swaptions for MTGE when compared to AGNC (proportionally speaking) and/or a higher overall weighted average negative duration.

With these slight differences noted, the current fixed-rate agency MBS valuations between AGNC and MTGE are still pretty similar. I am projecting there have only been minor valuation differences between AGNC's and MTGE's MBS portfolio through the first-half of the fourth quarter of 2013. As such, the points and data provided above for AGNC can generally be assumed for MTGE as well.

Conclusions Drawn:

By examining AGNC's MBS portfolio as of 9/30/2013, I have shown how and analyzed why management has continued to change the composition of AGNC's MBS portfolio. AGNC has continued to rebalance the company's portfolio holdings into slightly higher yielding coupons and continued to increase the proportion of its 15-year fixed-rate agency MBS. Generally speaking, 15-year fixed-rate agency MBS are less price sensitive to interest rate movements and help preserve BV in case interest rates rise at a fairly rapid pace. This strategy was directly shown via Table 2 above. Table 2 showed the quarterly net compositional changes that AGNC incorporated regarding its 15 and 30-year fixed-rate agency MBS holdings. During the third quarter of 2013, AGNC continued to increase its proportion of 15-year fixed-rate agency MBS holdings while decreasing its 30-year fixed-rate agency MBS holdings. AGNC was also able to "re-roll" its MBS portfolio into higher paying coupons within the same maturities. As such, a slight increase in yield should eventually come to fruition.

Through the week ending 11/15/2013, cumulative quarterly 15 and 30-year fixed-rate agency MBS price movements ranged from slightly negative towards to the lower-end of the MBS coupon spectrum to slightly positive towards the higher-end of the MBS coupon spectrum. As such, through the week ending 11/15/2013, I am projecting AGNC's MBS portfolio sustained a valuation loss of ($90) million. I am also projecting a similar slight valuation loss regarding MTGE's MBS portfolio (minor subtle difference in valuation).

Final Note: I would caution readers that MBS price movements change daily. As a direct result, AGNC's MBS portfolio has daily valuation fluctuations as well. One quick, sharp move in mortgage rates/U.S. Treasury yields can materially change MBS prices thus changing AGNC's MBS portfolio valuation by hundreds of millions of dollars. This is why I feel it is imperative readers understand how AGNC values its MBS portfolio and how modest changes in overall rates/yields have a direct impact on MBS prices (hence asset valuations). Therefore, I felt it would be extremely beneficial to provide readers a mid-quarter update regarding the valuation of AGNC's updated MBS portfolio as of 9/30/2013 (including projected current quarter activities).

PART 2 of this analysis will be out shortly. PART 2 of this analysis will analyze and discuss AGNC's derivatives portfolio as of 9/30/2013 regarding its composition. Following PART 2, PART 3 will discuss AGNC's derivatives portfolio regarding its valuation through the week ending 11/15/2013. PART 3 will also include a projected BV per share amount as of 11/15/2013.

Source: American Capital Agency's Mid-Q4 2013 Composition And Valuation Analysis - Part 1