Seeking Alpha

MeridianGuy's  Instablog

Send Message
Individual Investor and founder of Meridian Investments (a private investment club).
  • AGNC: Sustainability Analysis For The Layman (KISS) 31 comments
    May 3, 2013 5:11 PM | about stocks: AGNC, MTGE

    I am writing this post in hopes of getting a few responses to my thoughts on other SA posts concerning AGNC's recent earnings. I am long (AGNC) and (MTGE) and admittedly not a financial professional, however, I do have a Finance degree and been following stocks since I was 8 years old.

    Very often I feel like "intelligent" investors over-complicate their analyses and "laymen" investors get lost in the details, fast forwarding to the author's point/recommendation at the end of the article. This is extremely dangerous as many of these analyses have GAPING HOLES in them.

    To me the value of any business should be determined solely by the cash flows provided to the investor over time, not the next 2-3 quarters. This is a simple concept, but seems to escape most new investors I talk to.

    The first article I'd like to discuss is Scott Kennedy's Dividend Sustainability Analysis...

    This is a great article in terms of explaining how AGNC operates and why/how it pays its dividend. But as far as sustainability goes, there are two fundamental issues related to sustainability that are not addressed:

    1) The funding source of recent "Overpayments".

    2) The business model from a cash flow perspective.

    The first and most glaring issue I have with the article is using the company's history of overpaying as an indicator of the safety of the dividend. While this history may predict the likelihood of an "attempt" to repeat the practice through 2013, it has nothing to with
    the reality of it happening, or its sustainability long term. In fact, if anything this practice is UNSUSTAINABLE over the long term.

    Looking at his chart

    he provides the following info:

    Over Payment Chart
    200842m96m (225%)54m
    2009114m230m (200%)116m
    2010247m887m (360%)640m
    20111,030m1,530 (150%)530m

    Each quarter, they are paying out much more than the IRS requires (~230% vs. 90%). Most people compare each quarters dividend vs. that quarter earnings, but this is a mistake with REIT's. This quarter's earnings are driving NEXT YEAR'S payout requirements and should not be used to overpay this year's liability (IMO). Overpaying is just kicking the can down the road.

    I could make this really complicated and outline how they are achieving this through further capital raises, but in an effort to "Keep it Simple", my point is that their practice of overpayment REDUCES the sustainability of the dividend. If they were paying only what they owed, they would have a HUGE stash of cash in their UTI bucket (Undistributed Taxable Income) which would transparently show whether or not they would be able to pay next years dividend. Overpayments simply represent an early distribution of the current year's taxable income (technically it is not due until the following year). What I don't know is what happens if/when they cannot distribute 2013 profits because they are losing money
    in 2014?

    Which leads to my 2nd sustainability issue...

    As mentioned, I have a finance background, but am not a financial professional. With that caveat, this is my understanding of their business (Someone please correct me if I am wrong)...

    1) They raise 1B of capital.
    2) They are able to borrow 7-8x their capital in Short Term Notes.
    3) They borrow 8B at relatively low Short Term interest rates (say 2% and a 1 month duration).
    4) They use that 8B of cash to buy Long Term notes at higher rates (say 3.5% on 30 year mortgages).
    5) Each month they pay 2% interest on their ST debt and collect 3.5% on their LT investments.

    This 1.5% spread is "profit" so to speak. It does not sound like a lot, however, given that they are leveraged 8-1, this means they are getting 12% on the 1 billion of capital. The higher the spread, the higher the profits. 2% = 16%, 3% = 24%. The higher the leverage, also the higher the profits.

    ADDING COMPLEXITY I... SHORT TERM ROLLOVER RISK. (here's where it starts getting complicated)

    6) Each month, the company has to "rollover" its short term debt as it becomes due. So if ST interest rates go up to 2.1%, they have to borrow at 2.1% to pay off their 2.0% debt and their investment spread goes down to 1.4% (remember LT investments are yielding 3.5%). This ST vs LT interest rate risk is important to understand. In theory rates can go up .5% and they are fine from a cash flow perspective. They are still making a 3.5 vs 2.5 pct spread of 1 pct.

    The obvious disaster scenario is if short term rates spike 2% and their ST borrowing costs are 4% vs. an investment return of 3.5%. In this situation, they are losing money and losing it fast (remember they are leveraged). This is probably why their "risk" charts don't show rate increases larger than their spread. If that happens, the model is broken.

    7) When interest rates move, so does the value of their investment portfolio. If ST rates go up, so do LT rates, which means there are
    better LT investments available to outside investors, so those investors are willing to pay less for AGNC's 3.5% assets.

    Personally, I don't care about this devaluation of assets as long as I am still getting my spread. Unfortunately though, the IRS does care and treats these changes in valuation as income. In absence of Overpayments, this would not be too big an issue, but...

    In a decreasing interest rate market (like the past 4-5 years), note valuations increase, creating taxable income and committing AGNC to dividend liabilities higher than their cash flow. This is not an issue though as they can simply sell their LT investments at higher prices to cover the additional liability.

    As interest rates increase, the opposite happens, but as mentioned, this portfolio fluctuation isn't that important IF they are still making spread and not overpaying dividend liabilities.

    Given that (AGNC) is spending this year's earned cash by overpaying last years dividend liability, lets look what REALLY happens when interest rates rise...

    1) Investment spread goes down... Cash Flow goes down.
    2) Now current year cash flow is less than previous year cash flow.
    3) IRS demands (AGNC) pay 90% of previous years cash flow as dividends.
    4) (AGNC) doesn't have enough cash flow to cover dividend liabilities, so they have to sell LT investments.
    5) LT investments are lower than they were purchased for, so these "paper" losses become "Realized" losses.
    6) Smaller Investment portfolio = Lower Cash Flow.
    7) Downward spiral continues until rates drop/stabilize.

    In a nutshell that is my concern... the Overpayments are creating an UNSUSTAINABLE situation and forcing distressed asset sales. Eventually rates WILL rise, cash flow WILL fall, and they WILL be forced to sell investments at a loss to cover dividend liabilities. If they had never made an overpayment, their UTI balance would always have enough to cover the current year's dividend liability and the dividend would be easily forecasted (similar to Oil Pipeline's contracting their rates and volumes well ahead of time).

    In reality I think that when rates rise, instead of selling their LT investments, (AGNC) will just raise more capital to cover the dividend payments. BUT, remember they will have "paper" losses in an increasing interest rate environment. So even if they can cover current year's dividend payments (which they will by law), there will be no taxable income. And if there is no income, then no required dividends for the next year.

    Am I saying they won't pay dividends if they have paper losses for an entire year? No way. More than likely they will continue to have positive cash flow on the spread, but any new payout will be another Overpayment that will have to be addressed down the road. Also, they will likely continue to do capital raises to fund dividend liability, but what if the market tires of this model. Remember as LT rates increase, this vehicle becomes less appealing.

    One thing I may be missing is whether they can "Claim" previous overpayments against current year's liability. So if in 2013
    they make 3B in profits, but in 2014 they have only 2Bm of cash flow and 2B of paper losses, can they "claim" their 2008-2013
    overpayments to offset their 2014 dividend liabilities for the 2013 tax year? I don't know. In my experience logic and the tax code
    don't belong in the same sentence, much less the same article, and thus far I have not heard anyone say that is possible. I am hoping someone familiar with REIT tax laws can explain this to me, because this is may greatest concern. Remember, the forced selling of distressed assets is what got the banks in trouble and could easily happen here.


    AGNC's business model really isn't that complicated. They borrow at x rate and re-invest at (x + y) rate. As long as rates don't move more than their current spread (1.5%), VERY QUICKLY, they will make money. It really is that simple.

    That said, be prepared for major swings in the equity's valuation as earnings and dividend payments fly all over the place when crap hits the fan. Their overpayment situation has put them in a pickle and if they are forced to pay current year dividend liabilities that are higher than current year income, they may have to sell distressed assets ala banks 2008 and kaboom goes the stock. I really wish they limited dividend payments to actual liability. The spread-leverage model works and this vehicle exposes average investors to a financial instrument that was once only available to banks.

    I am still holding my (AGNC) and (MTGE), but am removing the DRIP option and considering dumping altogether in favor of pipeline stocks.

    Disclosure: I am long AGNC, MTGE.

Back To MeridianGuy's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (31)
Track new comments
  • Hi Meridianguy. First off, I don't know where you are getting your 2011 payout ratio from. Actual Net Income reported by AGNC for 2011 was $770.48 million. There was $171.315 million in unrealized losses which have to be added back to that to arrive at a reliable net income figure of $941.795 million, or $6.14/share. Taxable income was $1 billion.


    Total dividends declared for 2011 was $886.518 million, or $5.60/share. Actual cash paid out for dividends in 2011 was $663.506 million, per the cash flow statement. Again, I don't know where you got your numbers for 2011. AGNC clearly made more in Net Income, on both an adjusted GAAP and tax basis, than what they declared or paid out in dividends.


    For 2012, Net Income per the 10-K was 1.277 billion, less $10 million paid on preferred dividends leaves $1.267 billion. There was $580 million in unrealized losses from derivatives that have to be added back to that number, which yields $1.847 billion in Net Income, or $6.08/share. Taxable Net Income was $2.1 billion, per page 50 of the 2012 10-K. Common dividends declared were $1.518 billion, or $5/share. So, again, AGNC paid out considerably less in dividends than they earned for 2012. Total cash paid in dividends was $1.415 billion, per the cash flow statement.


    Based on this, I don't agree with your dividends paid or the payout ratio.


    I haven't reviews years prior to 2011 and 2012.
    3 May 2013, 11:04 PM Reply Like
  • Author’s reply » Hi Bryce, thanks for the responses. My figures were coming from the worksheet provided by Scott Kennedy's article. Not sure where he got those and is a great example of "Reader Beware". Looking at GOOG's numbers, they are different, but still reiterate my concern.


    My understanding is that "Dividend Liability" is driven by the PREVIOUS year's Net Income. The reason this is important is because if they face a year of losses (paper or real), they will have to raise funds to meet the dividend liability. This may mean raising additional capital, or selling distressed assets. I agree that regardless of "paper" income, they generally have positive cash flow, but using excess cash flow to "Overpay" dividend liability means they are spending money that is earmarked for future dividends. When that time comes, how are they going to finance those payments. See below (updated with GOOG numbers).


    YEAR - Net Income vs. Actual Payout vs. Required Payout (Dividend Liability)
    2013 - ??? vs ??? vs 1149
    2012 - 1277 vs 1415 vs 693
    2011 - 770 vs 664 vs 260
    2010 - 288 vs 173 vs 106
    2009 - 118 vs 80 vs ???


    Looking at this from a Payout ratio (dividends paid vs net income), you are right, they are fine. However, comparing actual dividends paid to what was owed (according to IRS), they are paying out much more. My question is what happens when Net Income and Cash Flow does not cover the previous years dividend liability. What if the next 3 quarters were like this quarter?


    Do you see my point? This all may be moot if they are allowed to claim their "overpayments", but my point is that we should be looking at their cash flow (real profits) vs. 10K Earnings which track mark to market adjustments.
    6 May 2013, 09:44 AM Reply Like
  • Author’s reply » Hey Bryce, I looked at Scott's numbers and it looks like he was adding in "Tax Book Differences". Those were the numbers I was using. But as mentioned, either way they are still overpaying and my concern is whether they can claim those overpayments against future dividend liability. For example, can they choose not to distribute anymore dividends this year because they have already overpaid by 1.4B over the past 4 years?

    6 May 2013, 10:43 AM Reply Like
  • "My question is what happens when Net Income and Cash Flow does not cover the previous years dividend liability. What if the next 3 quarters were like this quarter?"


    I suspect they would have some retained earnings, which may be in the form of cash or securities other than cash, but probably won't have a lot of that since they pay out so much in dividends. Other sources would be equity issues, debt, or selling of MBS. And they would probably reduce the dividend if income wasn't sufficient to pay the current dividend.


    I disagree that cash flows are "real profits". Cash flow includes all so called non-cash items, such as depreciation and premiums paid on MBS. However, cash was paid for the assets upon which depreciation and amortization is taken. The depreciation/amortization is simply the allocation of the cash expense over time. To say such expense are "fake" or not real is totally not credible. The company paid cash for the assets. So, I disagree that cash flows are the "real profits" as a result. They tell you how much cash is on hand or available to use to pay dividends and other liabilities/expenses, agreed, but they are not the true measure of the actual cash expended. Over time, accrual earnings are the better measure of your actual profits/earnings, and tell you how much you will have available to pay for dividends or other obligations.
    6 May 2013, 11:19 AM Reply Like
  • "I agree that regardless of "paper" income,"


    Sounds like you doubt the validity of accrual accounting and believe cash accounting would be better? I can assure you cash accounting is inferior to accrual accounting and can explain in detail why that is so. Maybe you can explain what you mean here.
    6 May 2013, 11:23 AM Reply Like
  • Author’s reply » I am probably over simplifying my definition of Cash Flow. As a "layman" I am thinking of it as net spread (after hedging/overhead). Their basic business model seems to be borrow ST and re-invest LT. When I say "Cash Flow" am referring to the returns on that model.


    In general, they should ALWAYS have positive cash flow so long as their avg ST borrowing rate is below their avg LT investment return. I am saying we should be focusing on how much this "core" business is producing on an annual basis.


    Scott's spreadsheet tracks "UTI - Undistributed Taxable Income". I assume this is what you are referring to as "Retained Earnings" and it is definitely a possible source for paying dividends. However, my thoughts are that the more you "overpay" in dividends owed this quarter, the less this bucket has in it for next years dividend.


    Personally, I would like to see UTI always have enough to cover the next year's dividend. I understand this would mean capital is going unused most of the year, but it would make me feel better because once the money is earned, it must be later distributed.


    To me it is no different than personal income taxes. It would be like us paying 2012 taxes during each 2013 paycheck, instead of withholding money each check for our 2013 liability. On an ongoing basis it is kind of the same (each paycheck has GI minus taxes), however, you will always have future unfunded tax liability if you are not withholding taxes related to that period's earnings. The IRS does not let you do that, and as an investor, I'd like to see AGNC withhold 90% of this quarter's earnings to guarantee next year's quarterly dividend.


    By not withholding their future dividend liability, they are dependent other forms of funding... additional profits, capital raises, or asset sales. If rates rise, profits fall, capital gets more expensive, and asset sales become realized losses.


    Do that make sense?
    6 May 2013, 12:01 PM Reply Like
  • Author’s reply » By "paper" income, i was referring to Mark to Market valuations. Ex:


    1. Raise 1B capital.
    2. Borrow 10B @ 2.5%.
    3. Invest 10B @ 4%.
    4. Collect 400m in investment interest.
    5. Pay 250m in borrowing costs.
    6. Add/Subtract 10-30m in hedging/transaction costs/profits.
    7. "Real" income = ~120m.


    If rates dropped 1% over this period, the 10B in investments might rise to 10.5B. This would be a "paper" profit of 500m. Or if they fell it could be a 500m "paper" loss.


    My point is that reported NetInc includes this "paper" mark to market amount. Meaning their "earnings" are -380m to 620m.


    I believe their "required" payouts are based on this -380/620 amount and not the "core" 120m of business profits. Is that true?


    Either way, I'd like to see what their "core" business is producing.


    I am an owner of a few rental properties. Some of which have gone underwater, however, my income still exceeds my borrowing costs and so I don't really care about my "book value". If my plans are to hold the assets and continue to collect income, then why do I care about book value?


    To me, the same should be true for AGNC. Of course my borrowing costs are fixed, whereas AGNC's is constantly changing, but my point is who cares about the fluctuation in BV?


    Also, in a increasing rate environment, aren't prepayments a good thing? In this case most assets are valued under face value so a prepayment (at face value) means the borrower is paying over market rates. I guess this assumes all assets are purchased at face. If they paid $107 on a $100 note, I can see how a prepayment could be bad.
    6 May 2013, 12:17 PM Reply Like
  • "Do that make sense? "


    It makes sense.
    6 May 2013, 12:48 PM Reply Like
  • "If rates dropped 1% over this period, the 10B in investments might rise to 10.5B. This would be a "paper" profit of 500m. Or if they fell it could be a 500m "paper" loss.


    My point is that reported NetInc includes this "paper" mark to market amount. Meaning their "earnings" are -380m to 620m.


    I believe their "required" payouts are based on this -380/620 amount and not the "core" 120m of business profits. Is that true?"


    I don't think that is accurate. The Mark to Market gains/losses are unrealized until the MBS is sold. My understanding is that only upon sale is a gain/loss "recognized" for tax purposes.


    Ideally, you want your "core" business to generate sufficient profit to pay the dividend, but I don't see that with Annaly or AGNC. I'm not that experienced with either, but it looks to me like Annaly comes nearer doing that than AGNC. Not counting realized losses on derivatives, I think Annaly generated enough income to pay their dividend in 2012 from their core business or from the net interest spread. But then they had a humongous loss on derivatives. I hope AGNC can get to the point of generating sufficient net interest income to pay for the dividend.
    6 May 2013, 12:52 PM Reply Like
  • Author’s reply » So it sounds like you are saying AGNC is not making enough in their core business to cover the dividend. This was my fear, but I don't see anyone talking about this in their Sustainability discussions. Is there a line item I should be looking at to see this number? My fear all along was that they are financing the dividend through capital raises, which is not much different than a Ponzi scheme. I hate using that word because I believe in their "core" model. In fact, the thing that has kept me invested is the idea that even if their "core" model currently only supports a $1 dividend, and that business is cut in half (.50 cents), you are still looking at a 6.5% yield with NO equity valuation reduction ($2/$30). Everyone (ie Cramer) says a 15% is an unsustainable dividend, so why does the price have to come down as the dividend comes down. Isn't a 6-8% dividend reasonable? It seems to be for MLP's so why not for REITs?
    8 May 2013, 11:00 AM Reply Like
  • "So it sounds like you are saying AGNC is not making enough in their core business to cover the dividend."


    That is true. See page 48 of 2012 10-K for the "Net Spread Income" calculation of $3.92 for 2012 and $4.66 for 2011.


    They have to make up the difference between $5.00 and $3.92 with gains on sales of MBS or derivatives, neither of which is sustainable in the long term. They had been able to make up the difference with gains on sales of MBS, but that didn't hold for 1st qtr. 2013, and no one has talked about why. There was a huge swing from $353 million in gains on sales of MBS in 4th qtr, 2012, to a ($26) million loss in the 1st qtr. 2013. AGNC did not address this in their conference call.


    They haven't been financing the divi through capital raises as I see it. For both 2011 and 2012 they had sufficient funds to pay the divi. Unrealized losses from derivatives used as hedges have to be added back. Not all of that should be because hedges aren't perfect in mitigating the risk, but a large portion of it should be, so I add all of it back. So, the GAAP numbers are almost always materially wrong, can't rely on them due to the large unrealized gains/losses on derivatives. Not many SA authors understand this. NLY corrects for it each quarter in their 10-Ks. Their reported GAAP number for 1st qtr was $.70/share, but after correcting for the unrealized amounts and a few other items, their adjusted number they reported was $.47/share.


    Some MLPs are ponzi schemes, like LINE. The only thing keeping it propped up is the confidence that LINE's reporting of DCF is right and Barron's is wrong. There are solid arguments for why LINE"s DCF is materially wrong. I believe it is and am looking to sell when it reaches $40/share again.


    The price of AGNC comes down due to BV and market forces, I would assume.


    I don't buy what Cramer is selling. He's wrong more than he's right. Start keeping a list of his picks and track them over time.
    8 May 2013, 11:29 AM Reply Like
  • I'd recommend looking at Realty Income Trust "O". It's overvalued now, with the market as high as it is, but looks to me like a stable, long term investment. The divis it pays have remained stable since it went public in 1996. Not one quarter the divi has declined, and it's increased over time. I don't understand the fundamentals of it. I rely on an SA author, Brad Thomas, to keep me apprised on it. Have made 40% on it since Nov. of last year. Will be looking to add to it when the market corrects, and invest in one more that Brad recommends.

    8 May 2013, 11:43 AM Reply Like
  • In regards to your analysis of their debt financing:


    "6) Each month, the company has to "rollover" its short term debt as it becomes due. So if ST interest rates go up to 2.1%, they have to borrow at 2.1% to pay off their 2.0% debt and their investment spread goes down to 1.4% (remember LT investments are yielding 3.5%)."


    They convert a large portion of the variable rate debt financing to fixed rate through interest rate swaps. They agree to pay a fixed rate of interest to a counter-party and the counter-party agrees to pay them a variable rate. This is a hedge against rising rates. As rates rise, the hedge is worth more to AGNC, since what they are getting paid increases with rising rates, but what they have to pay out remains fixed. So, this helps offset the rise in rates of their repos or short term debt financing. It mitigates the rise in costs of their debt.
    3 May 2013, 11:16 PM Reply Like
  • Author’s reply » Agreed. I was trying not to get too complex. My understanding is that the same hedge happens on the downside, so they don't get as much benefit as you'd expect when rates drop. The risk I was trying to highlight was ST borrowing vs. LT investing. In general they have short term contracts, hedged or not, and they will have to refinance/rollover those contracts. If rates rise fast enough, the average borrowing cost against their LT investment can turn negative (in theory). Do you agree?
    6 May 2013, 09:52 AM Reply Like
  • I don't know about that. Not experienced enough to know. However, if they have a swap contract, where they pay a fixed rate and receive a variable rate, it would seem to me that with a swift rise in rates that the swap would reset after a month, so that they are receiving more on the swap. But, again, not experienced enough to know how it all works out.
    6 May 2013, 12:57 PM Reply Like
  • Author’s reply » Ok, we are both out of our element, but my guess is they probably make hedging decisions based on probabilities and internal rate projections. And more than likely they don't want to spend money unnecessarily protecting against MAJOR moves that aren't likely to happen. In the options world, you may buy a put spread vs. a put in order to limit your hedge cost. Not sure if there are swap products like that. Where the swap writer has capped losses. Again, sounds like we are just guessing and hoping they are hedging appropriately.
    8 May 2013, 11:21 AM Reply Like
  • AS long as AGNC's taxable income each quarter exceeds the $1.25/share dividend, on a cumulative basis, I fail to see the problem you present concerning cash flows.
    3 May 2013, 11:19 PM Reply Like
  • Author’s reply » To me Net Income is meaningless. I am more concerned with Cash Flow from Operations.


    For example, they could have 1B of "rock steady" cash flow on 10B of investments. But if in one year their BV drops to 8B, they will show 1B of losses (1B profit - 2B paper losses). Because they really have 1B of "rock steady" cash flow, their dividend should be fine.


    My concern is the next year when their BV rebounds from 8B to 10B. Now all of a sudden they have 3B of profits (1B of cash flow + 2B paper profits). Now they will owe 2.7B of dividends. Where will they get that from if they are only making 1B in real cash flow?
    6 May 2013, 09:56 AM Reply Like
  • "For example, they could have 1B of "rock steady" cash flow on 10B of investments. But if in one year their BV drops to 8B, they will show 1B of losses (1B profit - 2B paper losses). Because they really have 1B of "rock steady" cash flow, their dividend should be fine."


    They would still show 1B in earnings, or Net Income, but show 2B loss in "comprehensive income". I put more weight on the Net than I do the comprehensive. And their taxes would not be on the 1B-2B =-1B loss but on the 1B Income, adjusted for non-taxable items.


    "My concern is the next year when their BV rebounds from 8B to 10B. Now all of a sudden they have 3B of profits (1B of cash flow + 2B paper profits). Now they will owe 2.7B of dividends. Where will they get that from if they are only making 1B in real cash flow? "


    I disagree with you here. Their taxable income would still be 1B adjusted for non-taxable items. The 2B comprehensive income would not be taxable. It has not been converted to cash. Only 90% of 1B would be due in dividends, not 2.7B.


    GAAP income and taxable are very different in some respects. GAAP recognizes accrual (non-cash) transactions, whereas tax accounting only recognizes cash transactions, with some exceptions like inventory.
    6 May 2013, 01:02 PM Reply Like
  • Author’s reply » Ok, thanks. That was another concern/misunderstanding. I wanted to verify exactly what drives their Dividend Payment requirements... ie - Does it include M2M adjustments or not. My impression from other articles was that it did.


    So would you agree the recent quarter's "important" number for LT investors is .78 cents a share (Net Spread Income) * 90% = .702 * 4 = $2.80 Dividend / $30 = 9.3% yield?

    8 May 2013, 11:49 AM Reply Like
  • I'm pretty new with AGNC. I do view the net spread income as important, yes. Not sure if dollar roll income is counted as taxable income or not, but they made $.40/share, I think it was, from that in 1st qtr., which helps pay the dividend. I have been looking for the 10-K to come out for 1st quarter, but haven't seen it yet.


    AGNC should state in its 10-K how much taxable income they had for 1st qtr.
    8 May 2013, 11:57 AM Reply Like
  • One more thing. There's a free program that will convert the 10-Ks, or any other documents, into PDF files, so you can type in search words to find specific information, so you don't have to reread a 10-K until you find what you are looking for. Saves a lot of time.

    8 May 2013, 12:21 PM Reply Like
  • Author’s reply » Thanks for the tip! I will check that out. The link I posted above is one of their 8k's. I haven't found the 10k, but this summary from the 8k seems pretty good...




    • $(1.57) comprehensive loss per common share, comprised of:


    ◦ $0.64 net income per common share


    ◦ $(2.21) other comprehensive income (loss) ("OCI") per common share


    ▪ Includes net unrealized losses on investments marked-to-market through OCI


    • $0.78 net spread income per common share


    ◦ Comprised of interest income, net of cost of funds (including interest rate swaps) and operating expenses


    ◦$ 1.18 per common share, including $0.40 per common share of estimated net carry income (also known as "dollar roll income") associated with purchases of agency mortgage backed securities ("MBS") on a forward-settlement basis through the "to-be-announced" ("TBA") dollar roll market


    ◦Includes $0.09 per common share of estimated "catch-up" premium amortization benefit due to change in projected constant prepayment rate ("CPR") estimates


    •$0.50 estimated taxable income per common share


    ◦Estimated taxable income was negatively impacted by net realized losses of approximately $(0.55) per common share recognized for tax during the quarter due to monthly settlements of TBA dollar roll positions during a period of price declines


    ◦Total estimated net taxable income (loss) attributable to TBA dollar roll positions was $(0.15) per common share, net of estimated TBA net carry income


    •$1.25 dividend per common share declared on March 7, 2013
    8 May 2013, 01:55 PM Reply Like
  • I would agree. While the fair value and book value of their MBS has taken a good hit, their earnings power still seems pretty good. I'll wait 'til the 10-K comes out to see what it looks like overall. I want to see what GAAP net income is, adjusted for unrealized gains/losses.
    8 May 2013, 02:37 PM Reply Like
  • Author’s reply » Agreed. As a W.A.G. I'd project a similar 1.18 in spread/roll income and then another 50 cent gain as a reversal of this quarters losses on the recent interest rate increase.
    10 May 2013, 03:43 PM Reply Like
  • Also, note: Taxable Income is not a reliable measure of a company's real earnings or profit. It fails to include some actual cash expenses of a company, then the timing of some revenues does not match the revenues with expense incurred to generate them. IRS regs were created to bring in revenue to the federal government, not to fairly measure a company's revenues or expenses, or Net Income. Tax basis accounting is not recognized by the Financial Accounting Standards Board (FASB) or the SEC as a legitimate basis of accounting.


    Tax basis accounting is a hodge podge of rules that have been stitched together over a long period of time by politicians and was never intended to be the basis for measuring the impact of financial transactions on a company's financial position or its operating performance. The 179 deduction and bonus depreciation are good examples of this.


    Cash basis accounting fails to record credit transactions of a company when they occur. It waits instead until cash is received before recognizing the transaction. Totally worthless and unreliable as a measure or performance.
    3 May 2013, 11:46 PM Reply Like
  • Author’s reply » I agree and is why I am attempting to look at Cash Flow vs. Net Income. All these conversations about Dividend Sustainability are looking at Payout Ratios which are based on NetInc which include Mark to Market valuations. My understanding is the IRS includes these fluctuations as part of their required payouts. Is that not true? Or are you saying the IRS is looking more at transactions on a cash basis and they company is not required to payout 90% of their 10k Net Income, but instead some other calculation?
    6 May 2013, 10:00 AM Reply Like
  • "My understanding is the IRS includes these fluctuations as part of their required payouts. Is that not true? Or are you saying the IRS is looking more at transactions on a cash basis and they company is not required to payout 90% of their 10k Net Income, but instead some other calculation? "


    The mark to market changes in value of the MBS are not a part of Net Income. They are included in "Comprehensive Income" but not Net Income. Therefore, these changes are not part of taxable income. No gain/loss has been realized on their sale as of yet. The income is counted only for GAAP purposes. However, an increase in value of MBS raises the book value of the MBS on the balance sheet, but the credit side of the transaction increases Owner Equity, as a separate section, and does not flow through the Income Statement at all. Same is true with a decline in value of MBS. When the MBS is sold, then a realized gain or loss is recognized, flows through the income statement, is taxable, and what has been recorded to Owner Equity as "Other Comprehensive Income/Loss" would be backed out at that time.
    6 May 2013, 01:10 PM Reply Like
  • "BUT, remember they will have "paper" losses in an increasing interest rate environment. So even if they can cover current year's dividend payments (which they will by law), there will be no taxable income. "


    Tax basis accounting is generally cash based, with some exceptions. So, generally paper losses or "unrealized" losses are not counted for taxes until they become actual or "realized" losses. I'm not aware that AGNC counts unrealized or paper losses for tax purposes. If so, I'd appreciate you showing me from their financials and from the tax code where that is at.


    The change in value of the MBS investments they hold are not considered for income tax purposes but considered "unrealized" until sold, neither are unrealized gains/losses from derivatives. See page 50 of the 2012 10-K for a reconciliation of GAAP Net Income to taxable income.


    "The primary differences are (i) unrealized gains and losses associated with interest rate swaps and other derivatives and
    securities marked-to-market in current income for GAAP purposes, but excluded from taxable income until realized or settled,"
    4 May 2013, 12:17 AM Reply Like
  • Author’s reply » This is really the same question I have and I was hoping someone would be able to answer this for me. It sounds like you are saying that M2M gains/losses are treated the same as they are for individuals.


    The reason i thought this might not be true is that AGNC is not actually taxed on their profits, right? My understanding is they get special treatment as their profits will be taxed on the investors personal taxes. So the "Tax Basis" argument is not relevant to AGNC earnings.


    What I am trying to understand is how the IRS determines how much AGNC must payout in dividends. I thought it was based on 10K Net Income, but I could be wrong.
    6 May 2013, 10:06 AM Reply Like
  • Mark to Market gains/losses or changes in Fair Value of MBS are "unrealized gains/losses" which do not impact Net Income. They are recorded as "Other Comprehensive Income/Loss" and go directly to Owner Equity, rather than Net Income/Loss. Therefore, they do not figure into "Taxable Income". I use the term Taxable Income the same as AGNC does. It's in their 10-K. AGNC is not required to calculate dividends payable on these unrealized gains/losses because they have not hit Net Income/Loss yet. Only upon their sale/disposal do the gains/losses become realized and are included in Net Income. At that point, they become part of taxable income and dividends must be paid on them.
    6 May 2013, 01:17 PM Reply Like
Full index of posts »
Latest Followers

Latest Comments

Posts by Themes
Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.