It always takes at least several seconds and sometimes a minute or two, depending on how fast Blogger is running. If you see one of these posts without a link at the bottom, just wait a bit and refresh and you will see it eventually.
"While you're at it, consider the non-agency (private label) mortgage market. Think west-coast (in particular) lenders with big Option ARM exposure. These guys are toast. WaMu, for example is trading today as if an FDIC takeover is imminent; the October $2.50 PUTs have traded over 10,000 contracts (as of 9:25 CT) against an O/I of 27,813 this morning - clearly, someone thinks they're a big fat zero between now and October expiration and is willing to shoot a decent wad at that bet." The Market Ticker
Thanks Tanta! That made sense to me...and while complex, not nearly as crazy making as, for instance, someone describing a derivative to me, or the alternative minimum tax...
Do individual loans move from the guaranteed to the retained in the event of default?
I obviously am incapable of explaining this to you, since I just wrote a big honkin' post on the subject and it didn't do anything for you. Maybe someone else can try explaining it again?
Tanta writes:
Do individual loans move from the guaranteed to the retained in the event of default?
I obviously am incapable of explaining this to you, since I just wrote a big honkin' post on the subject and it didn't do anything for you. Maybe someone else can try explaining it again?
Rather than gnash teeth over the niceties of legal/acceptable or descriptive terms versus terms of art I'm more concerned with whether holding for up to two years is good policy. In a period of declining values both the holding costs and depreciation are dramatically increasing the lost opportunity penalty. Just because they can hold up to two years doesn't mean they should. I also see a slippery slope in that they might find themselves end loading their workload. That would probably result in bulk disposals at even lower prices.
"What follows is my best effort to discover what the hell these people are talking about. I must disclose to you all that I am really just making an educated guess here."
and expecting them to use their brain as you would..
Tanta...I seem to recall a previous post (or coment maybe) that you made about them being able to send the loans back to the originators...in case of some structural defect or fraud on the part of the loan...am I making that up? If they can send them back, can they do that from both portfolios? I seem to recall further that there was a specific amount of time that they had to do this...and THAT is what I remember being 2 years...
The banks had research pieces out on this while ago, and none that I read were as lucid and straightforward as this piece. Thanks Tanta!
(Though in the end, I think they reached similar conclusions -- that this is doesn't change the nature of the business, it's simply a capital preserving move.)
OK, so this is all about Fannie's/Freddie's decision whether to exercise their repurchase option.
If they make that decision based on what will be cheapest in the long run, that is one thing.
But if they make the decision based on the desire to defer losses, that is quite another.
The accusation is that they have been leaving loans in the guaranteed portfolio which they know damn well are going to continue to be non-performing. So they continue to make payments to the bondholders, even though they will eventually have to repurchase those loans anyway... And they are doing this solely to avoid recognizing losses today. (Again, this is the accusation.)
Now, the IHT article you linked says:
The company had made decisions that, while not necessarily in violation of accounting rules, had the effect of overstating the companies's capital resources and financial stability.
Indeed, one person briefed on the company's finances said Freddie Mac had made accounting decisions that pushed losses into the future and postponed a capital shortfall until the fourth quarter of this year, which would not need to be disclosed until early 2009.
Is this not a reasonable layman's summary of the accusation?
This post is about financial accounting matters. If you are one of those people who drove us insane in the comments to yesterday's post on "assets" versus "liabilities" by arguing that "assets" are "really" "liabilities" because you, like Humpty Dumpty, are The Master, then you will find this post unsatisfactory.
Thank you. Most excellent Lewis Carol reference.
I am, however, willing to sell my liabilities to several people on yesterday's thread, at bargain prices. Just sayin'.
Nice post Tanta. Quite clear. They don't take losses for two years...
DUCKS!
My guess the real story is in the liabilities side. All that short term paper needs to role fairly frequently with lots of volume. IIRC, they had a big rollover at the end of September.
My guess only, can't find anything, is that the cost of rolling was too high. I'm trying to find something, haven't yet, but that's the only thing that makes sense here.
The only "two years" rule that matters is that the only GSE paper that's even arguably backed by the bailout is the two year maturity, and even this is implicit, based on the minimum time period expected of the conservatorship. Paulsen did not use the magic words "full faith and credit" . This is not Treasury paper, quite aside from the convexity issue, which is arguably worse given that the explicit aim of the program is to lower mortgage interest rates. The limits of this alleged guarantee seem to have been overlooked by many but not all commentators. See WSJ: HEARD ON THE STREET, "Future Shock for 'Frannie' Debt" By DAVID REILLY
September 9, 2008; Page C14
I'm more concerned with whether holding for up to two years is good policy.
The problem here is that your unwillingness to deal with the technical details leads you to use the term "holding" here in a way that makes NO SENSE.
Look, whether the loan is in the retained portfolio or the MBS, someone is trying to bring it current (collect) or work it out or foreclose. These things take varying amounts of time. Foreclosure rarely takes more than two years, but in totally weird and usual circs it can.
Nobody can refuse to "hold" the loan while these actions are taking place.
The point is merely that the MBS can "hold" the loan during the default servicing process only for up to two years. After that, Freddie has to "hold" it. Somebody has to "hold" it until it is reinstated or liquidated or whatever.
...we purchase loans from PCs when the loans have been 120 days delinquent and...(b) foreclosure sales occur...
How can they buy a loan after a foreclosure sale? Doesn't foreclosure wipe out the loan?
...(c) the loans have been delinquent for 24 months...
The loan has been delinquent for 120 days AND 24 months? Doesn't make sense.
As to the rumored change extending to two years the period before bad loans have to be reported, I understood that to mean the retained portfolio. But if Tanta says it isn't so, I'm a believer.
Apologies to tanta. why is the 2 yr situation any different from what Wells annoucned in its 2Q press release re delaying delinquencies to 180 days (I recall)? Merely a way to manage provisioning (smoothing function). Thuis also comes on heels of reports bacnks are simply delaying foreclosure. Also GSEs wrap has to buy back the defaulting mortgages in the pools and cure them. why would it be a surprise that they are delaying the repurchase as part of their wrap function? One would think in theory thsat the securities themsevels (pas throughs) would reflect the value of the NPL drag.
FYI posts from CRs vault which may or may not matter:
The Regulatory Reform Act also provides the Secretary of the Treasury with temporary authority, until December 31, 2009, to purchase any obligations and other securities Fannie Mae and Freddie Mac issue under certain circumstances. We intend to invest exclusively in Agency RMBS. The Regulatory Reform Act may lead to market uncertainty and the actual or perceived impairment in the credit quality of securities issued by Fannie Mae or Freddie Mac. This may increase the risk of loss on investments in Fannie Mae and/or Freddie Mac-issued securities. The Regulatory Reform Act could adversely affect the availability and pricing of Agency RMBS and, therefore, adversely affect our business. Welsh rarebit | Homepage | 09.06.08
A perfect example is SOP 03-3. Historically when mortgage underlying one of Freddie's MBS securities defauted Freddie would immediately buy it out of the trust, foreclose on it, and immediately book the loss. However, starting in Q1 2008 Freddie began to initially leave non-performing loans in the trusts hoping that they would cure. In effect, by delaying its purchase of non-performing mortgages out of the trusts Freddie began to delay its recognition of losses on the non-performing mortgages.GSE Guru | 09.06.08 - 7:25 pm
Fannie's Liquid Investment Portfolio (LIP) and Freddie's Liquidity and Contingency (L&C) portfolio are currently providing them with all the liquidity they need. Bear Stearns collapsed because it obtained most of its liquidity through overnight REPOs. The GSE's LIP and L&C portfolios currently do not have any REPOs at all! Unlike the major Wall Street firms Fannie and Freddie are in the business of holding securities until maturity. That means that as long as the mortgage delinquency rates do not get too out of line the value of the MBS securities they hold or guarantee should be irrelevant.
Anonymous | 09.06.08 - 10:21 pm
If they can send them back, can they do that from both portfolios? I seem to recall further that there was a specific amount of time that they had to do this...and THAT is what I remember being 2 years...
If the loan is defective (not just delinquent, but involving misrep) it can be put back to the seller/servicer from retained or MBS. And there is NO statute of limitations on that: the loan can be put back whenever the defect is identified.
It is rare that a 10-year old loan would be put back, because it's rare that a 10-year old loan that had performed for 10 years went bad and resulted in a loss. And defects are usually not found until the loan does go bad, and the GSE does a post-default quality control review. However, theoretically it can happen that way, because there is no time limit on the repurchase warranty.
Tanta, I am sure that you are right in your analysis. The principle is the same as a mortgage banker who finds himself obligated to repurchase a delinquent or otherwise defective loan from an investor opting instead to agree to sign an agreement indemnifying the investor against any losses incurred on the loan. Part of this indemnification would be advancing payments to reimburse the investor (servicer) for his advances on the MBS. The interest cost of the MBS advances is the coupon rate on the mortgage. Since the rate on the mortgage is fixed, one could easily have a defective low rate loan, say 5.75%, which becomes delinquent during a credit crunch where the mortgage banker's warehouse line cost could be considerably higher. For small mortgage bankers that could easily be the case, since warehouse lending agreements often call for higher rates and a much lower percentage of principal advance on repurchased loans.
On the other hand, the mortgage could have been originated during a time of relatively higher rates and if the time for repurchase comes after a Fed easing such as we have just experienced why would Freddie want to advance interest on a 6.75% loan when it had borrowing costs of 2.00%? Of course, this presumes that Freddie was capable of issuing more cheap "agency debt" and it would seem that recently that capability was called into question.
How can they buy a loan after a foreclosure sale? Doesn't foreclosure wipe out the loan?
All that really means is that the MBS never owns REO. By no later than the FC sale date, the MBS investor gets principal back.
The loan has been delinquent for 120 days AND 24 months? Doesn't make sense.
A loan could be 120 days delinquent for two years. That means that the borrower missed four payments, then started making payments again, but never caught up. That cannot continue for more than two years. (And it is never allowed to continue that long in the vast majority of cases.)
Mortgage-Related Security Cash Flows Because losses on sold loans are absorbed by the Enterprises directly and are not passed through to security holders, no additional credit losses are reflected in cash flows calculated for an Enterprises own mortgage-backed securities (MBSs) held as investments. Cash flows for single-class MBSs issued by an Enterprise and held as investments consist only of principal and interest payments. www.ofheo.gov/media/archive/docs/regs/ RBCFinal.pdf capital-restoration plans | 09.07.08
Fannie Mae has also created an accounting policy related to its MBS pool activities, which appears to have as a primary purpose the avoidance of consolidation under FASB Interpretation No. 46, "Consolidation of Variable Interest Entities," also known as FIN 46. OFHEO believes that in this area, the Enterprise has not complied with FIN 46. As part of its policy, Fannie Mae transferred securities between SFAS 115 categories, yet did not provide the required documentation to describe the rationale for the transfers. http://www.ofheo.gov/NewsRoom.as...e=Detail& ID=216
I also see a slippery slope in that they might find themselves end loading their workload.
You lost me with this one, Rob Dawg. Nowhere do they say or imply that they're going to instruct their servicers not to take any action on these loans for two years.
Many entities are in the process of implementing FASBs recently issued guidance on fair value, SFAS 157, Fair Value Measurements, and SFAS 159, The Fair Value Option for Financial Assets and Financial LiabilitiesIncluding an Amendment of FASB Statement No. 115. Proper implementation can require CPAs to exercise professional judgment. This professional judgment is being put to the test in the measurement of fair value for mortgage-backed securities (MBS) because of significant declines in values and reduced levels of trades resulting from the recent credit market crises. This article illustrates the application of this professional judgment to MBSs as well as a discussion on other-than-temporary impairments of MBSs under SFAS 115, Accounting for Certain Investments in Debt and Equity Securities. In addition to providing a primer on the accounting treatment of MBSs, the discussion of SFAS 157 and SFAS 159 is expanded to help CPAs apply these standards in other situations
An important exception exists for available-for-sale and held-to-maturity securities that have experienced an other-than-temporary decline in value. The exception, discussed below, requires that the unrealized loss be reported in the determination of net income rather than as a component of other comprehensive income.
The 2 years is simply the end of the option period. During the two years, Freddie can opt to either advance the payments to the investor at the mortgage MBS rate or buy the loan principle if Freddie's cost of funds is cheaper. They are responsible for the foreclosure losses regardless. The only issues are capital cost and cash flow. Until recently cash flow for GSE's had never been a problem.
The issue with Gretchen and other similar reporters is that they are writing about things that neither they nor their readers have any understanding about and they consistently get it wrong. They appear to have an agenda and anything they hear is filtered according to that agenda.
This post is a perfect example of why I no longer bother to read any articles in newspapers anymore, except possibly the headlines in order to be aware of what others will say or be thinking on the topic.
You know Tanta, I think there was a conspiracy to destroy these great organisations. Remember it started in '03 for Freddie and '04 for Fannie.
What was the big deal about > 11 billion in "accounting errors"?
P.S. In '03 and '04 it took real talent to screw up in real estate.
P.P.S. If my memory serves me well, Enron's "accounting error" was ~1/2 billion. I don't have to remaind you what happened to Enron's boyz and what happened to F&F's boyz.
"What shows up on Freddie's balance sheet is the "guarantee asset," which is the fair value of the guarantee fees received, and the "guarantee obligation" (over on the liability side) which reflects the fair value of the projected credit losses."
Haven't been following closely, but are you saying there was no regulatory capital requirement for the guarantee book?
Disclosure requirements may discourage entities from violating the intent of the fair-value election. An entity must disclose its reasons for electing the fair-value option. If a partial election is made, an entity must disclose the reasons for the partial election.
If it is probable that the investor will be unable to collect all amounts due according to the contractual terms of a debt security not impaired at acquisition, an other-than-temporary impairment shall be considered to have occurred.
(PCAOB) emphasizes the importance of determining an entitys principal market when pricing services are used for fair-value measurements The alert states that a pricing service might provide an amount for which a large financial institution could sell the financial instrument. However, a company that owns that financial instrument might not be able to transact in the same market as a large financial institution. If the price available to a large financial institution would not be available to the company, then that price may not be an appropriate measure of fair value under SFAS 157.
The use of Level 3 inputs is necessary when observable inputs are not available. For MBSs, this could happen when very little or no market activity is observed for identical or similar assets at the measurement date. In using Level 3 inputs, SFAS 157 (para. 30) indicates that the fair-value measurement objective remains the same, that is, an exit price from the perspective of a market participant that holds the asset. Assumptions in a model used to measure fair value should take into account assumptions that market participants would make in pricing an asset if the information can be obtained without undue cost and effort.
The current environment of foreclosures, defaults, and uncertain real estate values suggests increased risk and uncertainty of future cash flows for investors holding MBSs. Accordingly, measuring risk for some MBSs may have become more difficult.
As noted, fair-value measurements based on the use of unobservable inputs (Level 3) requires the exercise of significant professional judgment. Consistent with the subjectivity of these measurements, extensive disclosures are required by SFAS 157 (see paras. 3235). Among other disclosures, entities must disclose the levels (1, 2, or 3) in which inputs to fair-value measurements fall.
Re: Some of our financial instruments, such as our trading and available-for-sale (AFS) securities and our derivatives, are measured at fair value on a recurring basis in periods subsequent to initial recognition. We measure some of our other financial instruments at fair value on a non-recurring basis in periods subsequent to initial recognition, such as held-for-sale mortgage loans. Table 1 presents, by balance sheet category, the amount of financial assets carried in our condensed consolidated balance sheets at fair value on a recurring basis and classified as level 3 as of June 30, 2008. We also identify the types of financial instruments within each asset category that are based on level 3 measurements and describe the valuation techniques used for determining the fair value of these financial instruments. The availability of observable market inputs to measure fair value varies based on changes in market conditions, such as liquidity. As a result, we expect the financial instruments carried at fair value on a recurring basis and classified as level 3 to vary each period.
Excellent summary. I'm new here, and would appreciate a preliminary clarification -- is Tanta's blunt persona part of her schtick? More to the point, while everything in Tanta's post makes sense, why is Nemo's comment above not still relevant?
Most important, an entity can make the election on the adoption of SFAS 159 for financial assets and liabilities owned or issued before that date (e.g., on January 1, 2008, for calendar yearend companies). Additionally, the election can be made on a security when the security is purchased. Paragraph 9 of SFAS 159 includes a complete list of triggers allowing for an election. Once SFAS 159 is effective for an entity, the effect of the first remeasurement to fair value will be reported as a cumulative-effect adjustment to the opening balance of retained earnings (para. 25).
jkr0sc0 writes:
Excellent summary. I'm new here, and would appreciate a preliminary clarification -- is Tanta's blunt persona part of her schtick? More to the point, while everything in Tanta's post makes sense, why is Nemo's comment above not still relevant?
jkr0sc0 | 09.09.08 - 12:17 pm | #
I may not understand Nemo's comment. It seems to involve the question of whether Morgenson and Duhigg accurately described an inaccurate accusation or not. I, personally, simply don't care about the answer to that question.
The "accusation" is that Freddie's repurchase policy has the sole purpose of delaying recognition of losses. I just wrote a long post suggesting that the purpose is more likely to reduce funding costs and hence reduce the ultimate losses to Freddie. I don't care how faithfully these reporters stenographed the accusation. Since they didn't follow it up with trying to find out if it's true or not, they don't get credit from me for simply repeating unsubstantiated claims.
I t is funny how thousands of banks are fucking up accounting and cant understand this basic accounting shit that average person should be able to understand, like Gretchen, Paulson, Fuld, Dodd, Tanta...
It appears Freddie is tightening the criteria under which they transact guaranteed mortgages into held mortgages. Morganson mistakenly thinks they are tightening the criteria under which Freddie is recognizing impairments.
The one plausible issue is that Freddie's internal procedures likely call for a more thorough review of valuation upon transacting mortgages to "Held". On the other hand, the inputs into valuation on "Held" portfolio would probably be more on the Level 3 end which are more judgemental and less empirical.
Would it be fair to say the the increased cost of capital for F&F had a side effect where MBS repurchases ( and the immediate write-down that does along with it ) were put off.
This could be perceived as an attempt to obscure the true cost of the market change by keeping them in the "guarantee portfolio" longer and delaying the inevitable write-down as market conditions deteriorate.
Haven't been following closely, but are you saying there was no regulatory capital requirement for the guarantee book?
No, I am not saying that.
They have to hold capital commensurate to their entire risk as an entire business. I'm not going to get into the exact capital requirements today.
It is simply that you calculate "risk" differently on the two sides of the business. In the retained portfolio you own the loans. You have funding costs that can--and apparently have, recently--exceeded the interest rate on the loans, which gives you a negative interest margin and blows a big hole in your income.
That is a risk you do not have in the MBS portfolio, because your investors have it for you. It's their principal outstanding, not yours. You just pass through interest at the coupon rate.
In both parts of the business you have credit risk.
Well, after yesterday, I now know the difference between liabilities and impaired assets. Ms. Wikipedia let me find some definitions and Mr. Google led me to a 10-k for Fannie. They have a description of their business activities that sounds similar to what Tanta just posted.
I only hope this conservatorship action works out better than some of the other dramatic actions the government has taken the last few years. Otherwise, see the remarks from FFDIC above.
This could be perceived as an attempt to obscure the true cost of the market change by keeping them in the "guarantee portfolio" longer and delaying the inevitable write-down as market conditions deteriorate.
Or, it could be perceived as finding the least expensive way to hold delinquent loans while servicing is running its course: invest your own (borrowed) capital, or let the bondholders continue to invest capital.
Now that we the peeps are the conservators of Freddie, which do you think we should make them do? Pay off bondholders faster than they have to and increase costs to us, or let the bondholders carry the loans until foreclosure?
How many times do I have to point out that the guarantee program takes losses, too? "Delaying" showing a loss in the retained portfolio just means that it continues to show up in the guaranteed portfolio (as an increase in the guarantee obligation).
"I'm new here, and would appreciate a preliminary clarification -- is Tanta's blunt persona part of her schtick? "
IME, Tanta is impatient with sloppiness. She is particularly impatient with people who don't read what she wrote and then demonstrate it by asking questions that were just answered or making statements that were debunked in her post. She is actually quite patient and helpful with questions, even dumb questions, in other circumstances.
Do you have any comment on the tax asset criticism? This was also mentioned in a Bloomberg article.
Tax assets in my view are equivalent to deferred cash inflows (offsetting future taxes), albeit contingent on future profitability.
What I see in these criticisms is a tendency to associate capital with these specific assets, and an inability to get cash out of them when needed. But this is a false association in that it confuses the aspect of capital and liquidity. Write-offs and lossses don't necessarily generate a requirement for cash directly.
I'm also confused by the comment that financial institutions are not allowed to carry tax assets, which I've also seen in several places. If that's true, how do they square the balance sheet with the income statement when the after-tax cost of losses is less than the pre-tax cost?
"How many times do I have to point out that the guarantee program takes losses, too? "Delaying" showing a loss in the retained portfolio just means that it continues to show up in the guaranteed portfolio (as an increase in the guarantee obligation)."
Only if the inevitable future write-downs are properly documented. This is especially true with our current market where there is virtual certainty that the underlying asset is loosing equity every day ( the write-offs are only going to get worse ).
Don't get me wrong, I agree that not exercising their option makes the most sense today.
The "accusation" is that Freddie's repurchase policy has the sole purpose of delaying recognition of losses.
No, the accusation is not about the policy. The accusation is that F&F have lately been misusing their repurchase option specifically to delay recognition of losses, and thereby exaggerating their capital cushion.
Quoting from the IHT article again:
Indeed, one person briefed on the company's finances said Freddie Mac had made accounting decisions that pushed losses into the future and postponed a capital shortfall until the fourth quarter of this year, which would not need to be disclosed until early 2009.
IHT is citing a source who was "briefed on the company's finances", in the context of Morgan Stanley delivering its findings. I am not sure on what basis you say the accusation is "inaccurate", when it appears to be a simple and accurate layman's description of what Morgan Stanley (apparently) discovered when they reviewed the companies' finances.
Straight from Freddie's 2007 Annual Report.(page 27-28)
Minimum Capital. The minimum capital standard requires us to hold an amount of core capital that is generally
equal to the sum of 2.50% of aggregate on-balance sheet assets and approximately 0.45% of the sum of outstanding
mortgage-related securities we guaranteed and other aggregate oÅ-balance sheet obligations.
There are other risk based capital requirements. It seems they look at minimum, critical, and risk-based and take the highest requirement.
There are a handful of people that read your posts like this and continue to tell you that you are right.
The rest of the world doesn't 100% understand what is happening but at least investors are smart enought to understand the common is basically worthless.
Mortgage rates have lower a little bit but in time they will be head higher.
Tanta, you've made a very good explanation for why a company would do what they did.
But the really interesting question is:
When did Freddie start leaving the mortgages in securities longer? Did it correspond to when their cost of capital went above mortgage interest rates (making it an economically sound decision), or was it when capital became constrained?
Only if the inevitable future write-downs are properly documented. This is especially true with our current market where there is virtual certainty that the underlying asset is loosing equity every day ( the write-offs are only going to get worse ).
You are still confused.
Here's a classic example of how this works: an MBS loan goes 90 days delinquent. The servicer starts FC. FC takes another six months. That means that the total time this loan has to be held by somebody while it is delinquent is 9 months.
Freddie has an option: it can buy the loan out of the pool at 90 days down, meaning that it holds the loan in the retained portfolio during the six months it takes to FC. Or, it can leave the loan in the MBS during the six months it takes to FC. Either way, recoveries in FC are not affected: the liquidation proceeds are what they are. This policy does not increase the time it takes to liquidate loans. Therefore this policy does not increase credit losses.
Tanta, I actually liked the post and could basically understand it.
Just to set the record straight about yesterday, I never said that assets = liabilities or any such nonsense. I merely said that there was no way JSP (ie the Treasury) was going to wind up WITH the assets, he will wind up with the liabilities only.
I realize that there are some quite good assets in there, JSP just won't be getting any, that's all.
When did Freddie start leaving the mortgages in securities longer?
From the 10-K thing I quoted in the post:
Through November 2007, our general practice was to purchase the mortgage loans out of PCs after the loans became 120 days delinquent. Effective December 2007, we no longer automatically purchase loans from PC pools once they become 120 days delinquent,
I'm also confused by the comment that financial institutions are not allowed to carry tax assets, which I've also seen in several places. If that's true, how do they square the balance sheet with the income statement when the after-tax cost of losses is less than the pre-tax cost?
If that is true (and I've never done accounting for financial institutions), then one would have an allowance on the balance sheet (and an expense on the P&L), that would exactly cancel out the asset. This is often done with startups that may never be profitable, or won't be profitable without significant new capital (and the changes in ownership that come with new capital). There should be a Note accompanying the financials explaining the treatment of deferred taxes.
Did it correspond to when their cost of capital went above mortgage interest rates (making it an economically sound decision), or was it when capital became constrained?
Anonymous | 09.09.08 - 1:03 pm | #
I'm a bit confused here--what's the difference between capital being constrained and capital being expensive?
This reminds me of an esteemed Congressman I saw interviewed (CNN?) a month or so ago, making the claim that the GSE's have an advantage as they don't have to mark their portfolio to market (and this wasn't meant as an accounting advantage, but this Congressmen thought this would mean less losses in fact), as they can hold their loans to maturity. Ignoring for the moment whether or not this is an "advantage", though clearly the GSE's have exploited this accounting loophole (under the theory, presumably, that there is no established market for mortgages and so they carry at book value, ignoring the fact that there is a huge market for mortgages and that in fact the market treat mortgages as more-or-less fungible, at least enough so that it would not be difficult to establish a market avlue for the GSE portfolios, but I digress), the fact is you can't hold a mortgage to maturity when it defaults and forecloses and the entire reason MBS's from private-label issuers are trading at a massive discount to par is the expectation of these defaults and foreclosures.
In other words the only "advantage" the GSEs have is they get to delay recognizing losses, not that they avoid them.
But these kinds of accounting rules (if I hear one more person tell me about how "transparent" our financial system is and how great US GAAP is, I think I'm going to explode ) is why relying on financial statements in the US market is absurd.
I haven't studied the new international accounting rules in detail but I sure hope they are less prone to fraud and manipulation than US GAAP.
You have the general gist of what is occurring pretty spot on, however there is correction/distinction I would like to make.
You state "What you see on the balance shee is a guarantee asset"
Note that on the balance sheet, the only assets listed are the assets they actually own. (I.E. total assets = 800 billion)
These guarantee assets are actually held in a trust, that is completely off the balance sheet. (Although I do notice a guarantee asset,net on their balance sheet which I'm assuming is net of the liabilities, so I could be wrong here. However your general point about the different accounting rules between the classifications is correct regardless)
I.E. the reason they want to keep delinquent loans in the "guarantee" pool is because it's not on their balance sheet, thus they can hide the write downs.
This would be fine if they always re-bought the loans and put them on their balance sheet after 90 days delinquent, but delaying consoloditing these assets and marking them down is how they are smoothing or hiding their losses.
I was looking for the breakdown of these guarantee assets yesterday, and I noted on their 10-K it stated "Per the special SEC accounting rules, we do not have to disclose information about these off balance sheet guarantees. However, we have done so at www.fanniemae.comdisclosure"
Then I went to that website, and it's like a maze, and found nothing useful.
Furthermore, in thinking about this "guarantee asset"
the only way this could be created is if they spent cash on this, and then as they receive cash back they recognize revenue, which doesn't make sense from my understanding of the ways guarantees work.
Or that when they created this asset they recognized revenue, and then decrease the asset when they receive the cash.
So this 11B in guarantee fee "assets" could very very easily be a complete fabrication of revenue earned in the past. If they have lax standards on writing these "Assets" down, the overstatement of their earnings could be staggering, to the degree that they've actually lost on these guarantee payments (Since it's net of liabilities, if it's an actual liability, the earnings overstatement could be tremendous)
--2 years is the limit on how long Fannie/Freddie can delay repurchasing nonperforming loans. This doesn't mean they won't repurchase any before 2 years, it's just their outside limit.
--All mortgages they purchase are technically assets (even if they are nonperforming ones), while the MBSs they issue to investors("PCs" in GSE lingo) are liabilities.
This is a bit longer than 25 words, but I essentially conveyed the same information as Tanta's 3000 word essay in fairly short order, did I not?
Whyizzit we(U.S.) only take over the loser banks after they hit bottom? I say we take over a few good banks to offset the bad banks and take back the 28 mil we just paid to frannie's ceo's. Where's the accountability?
These guarantee assets are actually held in a trust, that is completely off the balance sheet. (Although I do notice a guarantee asset,net on their balance sheet which I'm assuming is net of the liabilities, so I could be wrong here
Look, I intentionlly used Freddie Mac's terms for these things so that if you wanted to you could find them on the balance sheet. If you look at the consolidated balance sheet from the last Q, under "assets" you find this line for "Guarantee asset, at fair value" at $11 billion. Pop down to "liabilities," and you see this line called "Guarantee obligation" at $14 billion. The difference between those two numbers means that credit loss-related costs on the guarantee portfolio are larger than the value of the guarantee fees collected. Yuk. I never said they weren't taking losses, remember.
Notice that under liabilities you also see this "Reserve for guarantee losses on Participation Certificates." That would be a function of the fact that the guarantee obligation is larger than the guarantee asset.
Putting aside the normative question of whether entities should legally minimize regulatory capital requirements (especially if they're publicly announcing the way they will be doing so), I'd second Tanta's core critique -- that the authors of the IHT article appear to have incompletely understood and investigated their own accusation. To me (and seemingly to Tanta), the clincher is the reference in the article to the companies injecting their own capital into pools of securities containing these loans -- as Tanta points out, such injections are the very thing F&F are trying to avoid.
If I understand Tanta's responses correctly, she's not closing the door on the possibility that F&F undertook the change in their repurchase policy due to accounting/capital requirement benefits, although she does appear to be asserting that economics, not accounting/capital requirements, was the driving factor. But she doesn't want to give credit to reporters who simply parrot accusations without adequate comprehension and/or follow up.
The two things still missing for me: (i) do increases in the guarantee obligation roughly track the write-down that would occur upon purchase of the loans by F&F and (ii) even if losses are roughly equivalent under either silo, do capital requirements ultimately differ (although I'd need a primer on capital requirements to even begin to parse the latter -- I don't think it's as simple as using the percentages 3 cents so helpfully posted)?
You have funding costs that can--and apparently have, recently--exceeded the interest rate on the loans, which gives you a negative interest margin and blows a big hole in your income.
All mortgages they purchase are technically assets (even if they are nonperforming ones), while the MBSs they issue to investors("PCs" in GSE lingo) are liabilities.
Well, no.
The MBS loans are off-balance sheet. They are therefore neither assets nor liabilities on the balance sheet. They are assets of the securitization trust.
What is on Freddie's balance sheet is its financial guarantee of those MBS.
The accounting issues are a complete red herring -- these 'issues' have been well known, nothing new, and nothing shady. Ditto capital ratios.
After all, Lockhart presided over pledging the books clean, and also presided over lifting the caps and reducing the capital ratios! And just in the last few weeks, Lockhart, Bernanke, et al all said there were no capital adequecy problems at the GSEs.
And yet, now, suddenly, Paulson has to move quickly to place the GSEs into conservatorship -- and he places Lockhart and Lockhart-selected people in direct control of the GSEs! If these new accounting problems were are real issue, why continue with Lockhart?
Obviously, a red herring.
But what I really want to know, is why isn't this aspect being discussed in the (free) press?
An off-topic question: I think I understand correctly that Fannie and Freddie invented mortgage backed securities -- and basically securitization itself. But we they also the founders of CDO style (multi-tranche) securitization or was that an innovation that they adopted from the private sector?
My very profound thanks for your willingness to share your expertise with us on this blog.
The MBS loans are off-balance sheet. They are therefore neither assets nor liabilities on the balance sheet. They are assets of the securitization trust.
What is on Freddie's balance sheet is its financial guarantee of those MBS.
I stand corrected --thanks for the assist! Is it any wonder the mainstream press or public gets confused about this stuff, when regular readers of this blog can't even get it 100% right?
Personally, I think complexity, opacity and obfuscation (ex: hiding liabilities "off-balance sheet") are a built-in feature of mortgage securitization. Banksters know that regular investors and taxpayers have no comprehension of or patience for this kind of stuff, and that's how they pull off these shenanigans and get away with it.
(i) do increases in the guarantee obligation roughly track the write-down that would occur upon purchase of the loans by F&F and (ii) even if losses are roughly equivalent under either silo, do capital requirements ultimately differ (although I'd need a primer on capital requirements to even begin to parse the latter -- I don't think it's as simple as using the percentages 3 cents so helpfully posted)?
Well, that is certainly where this all gets very difficult. Remember that any loan that is modified has to be bought out of the MBS when it is modified. That has not changed. So presumably they look at a net present value calcuation (losses in FC vs. loss if you modify, which automatically means you take it to the retained portfolio), and go with "less loss." I think that is probably what the reporters were trying to get at by saying "The companies have injected their own capital into pools of securities containing these loans, arguing that their new policies are helping more borrowers." I have no idea how you could do the relative calculation of cost/benefit if the loss calcs between the portfolio and the guarantor program were just totally incommensurate. But they are doing a lot of mods, so we have to conclude that in some cases they'd rather take the immediate write-down to the retained portfolio than to foreclose and let the MBS carry the loan until liquidation.
It is always difficult with the outfits formerly known as GSEs to know when this is an economic decision, when it's a capital/regulatory decision, and when it's that "mission" thingy. Nobody ever puts a line on the financials for "costs incurred in keeping Congress happy by preventing unnecessary foreclosures" or whatnot.
I don't assume that the Fs have ever done anything out of absolutely "pure" motives. I am just never quite sure what people mean when they talk about things being done "only for accounting purposes."
Thank you Tanta for pointing out the inaccuracies and spin put out by the media, but is there a primer available somewhere for what went wrong at the GSE's for those lacking higher education, let alone an accounting background? I had tried reading that 10-K on Sunday;, and while today's post helps a great deal, I still have newbie type questions that I'd rather not ask if answered elsewhere.
I'm just asking for directions to elsewhere. thank you
What's the bottom line at the end of the day? That's really THE issue isn't it? What increase of non-performing mortgages are we talking about and what $ amount is now needed?
How MUCH capital would now be needed if they were to take the guarantys onto their investment portfolio, regardless of the relative expense - advantage or disadvantage to funding costs?
How much are the losses if they were to simply assume the issues NOW?
You do a great job by the way. My question is that whether you're financing $100bn at 5% or 6% the rate isn't as important as the amount...
Re:hese transactions create what are called mortgage-backed securities (MBS). Fannie and Freddie only guarantee to the holders of MBS that the principal and interest on the mortgages in the trusts will be paid in full and on time. In other words, Fannie and Freddie take only the credit risk on these mortgages; the interest-rate risk--that interest rates will rise and the mortgages will become less valuable, or that interest rates will fall and the mortgages will be prepaid and disappear--is taken by the holders of the MBS. For their guarantee, Fannie and Freddie receive an annual fee as long as the pools remain in existence. In this segment of their business, Fannie and Freddie have guaranteed about $2.2 trillion in mortgages.[1]
In the other part of their business, Fannie and Freddie buy mortgages from originators and hold them in portfolio; they also repurchase for their portfolios some of the MBS that they have already issued. Because they are then the owners of these mortgages and MBS, they have assumed the credit risk, and because they have to borrow the funds to buy the mortgages and MBS, they assume the associated interest-rate risk. Of the two, interest-rate risk is far more significant, and Fannie and Freddie have to enter into large hedging transactions to mitigate it. They assume these substantial risks, and contract for hedging, because their portfolios of mortgages produce by far the greater part of their profits. This is true because, as GSEs, they are accorded favorable rates in the capital markets, and thus can arbitrage between their borrowing costs and the rate they will receive on the mortgages and MBS they hold.
I remember hearing a little while back that F or F was looking into requiring buy backs of mortgage paper sold to F and F where there essentially failure to comply with representations by the originator regarding conformity with F & F underwriting guidelines or early payment defaults.
Since you are very familiar with the way F and F does their accounting, my question is do these kind of contingent claims show up as any kind of significant entry anywhere on either of the F &F balance sheets, and if so, are they treated differently for portfolio vs. securitized loans?
I meant what the Q calls "the expected cost of holding the nonperforming mortgage in our retained portfolio."
i see. i think using the term "funding costs" is somewhat confusing, because it's not really about the cost of issuing agency debt (which is always substantially cheaper than the coupon on MBS) to fund the mortgage, and it doesn't really have anything to do with interest margin.
per Freddie:
Freddie Mac believes that the historical practice of purchasing loans from PC pools at 120 days does not reflect the pattern of recovery for most delinquent loans, which more often cure or prepay rather than result in foreclosure. Allowing the loans to remain in PC pools will provide a presentation of its financial results that better reflects Freddie Mac's expectations for future credit losses. Taking this action will also have the effect of reducing the company's capital costs. The expected reduction in capital costs will be partially offset by, but is expected to outweigh, greater expenses associated with delinquent loans.
do these kind of contingent claims show up as any kind of significant entry anywhere on either of the F &F balance sheets
You won't find that on the balance sheets anywhere. Put-backs of loans generally only happen when the loans have defaulted, so the put-back then counts as a "recovery" of losses already realized, I believe. If the repurchase volume is significant, there should be a discussion of the matter in the footnotes in the "credit risk" section.
How MUCH capital would now be needed if they were to take the guarantys onto their investment portfolio, regardless of the relative expense - advantage or disadvantage to funding costs?
From the last 10-Q:
For the six months ended June 30,
2008, we purchased approximately $1.1 billion in unpaid principal balances of these loans with a fair value at
acquisition of $0.9 billion. Although the volume of these repurchases has decreased in the six months ended June 30, 2008, there was $12.5 billion unpaid principal balance of loans remaining in our PCs and Structured Securities as of
June 30, 2008 that were greater than 120 days past due for which we have not exercised our repurchase option.
So the answer, at the moment, seems to be about $12.5 billion less the fair value adjustment, whatever that is.
Thanks for that. I'm assuming you are referring to FNM...So a 10 fold increase in non-performing or delinquent mortgages....and added to the tax credit of $20.6bn takes a total of $33.1 bn from their regulatory capital with this calculation...accounting for 70% of the $47bn, leaving just $14bn to cover $800bn in their investment portfolio and $2trn some odd in guaranty book?
Tanta wrote "How MUCH capital would now be needed if they were to take the guarantys onto their investment portfolio, regardless of the relative expense - advantage or disadvantage to funding costs?
So the answer, at the moment, seems to be about $12.5 billion less the fair value adjustment, whatever that is."
No, no, no. The $12.5 billion is the face value of the loans not the capital required. Assuming an 8% capital requirement (generally applicable to banks) that would translate into $1 billion in capital required. However banks (don't know about Freddie or Fanny) usually have a capital charge associated with guaranties or off-balance sheet items. Therefore, the additional capital required to bring the $12.5 billion in loans on-balance sheet could be less than the $1 billion.
No, no, no. The $12.5 billion is the face value of the loans not the capital required.
Sorry, I'm really getting tired.
$12.5 billion is what Freddie would have to pay the bondholders (par on UPB). $12.5 billion less the fair value adjustment is what would hit the retained portfolio (the difference being the write-down).
As you say, the capital reserve requirement would be less than that (if not 8%). But I thought Ed was asking a different question.
Ed, I'm talking strictly Freddie today. And remember that the retained portfolio balance is net of loan loss reserves. I'm getting too tired to look it up again, but I just looked this morning and I believe that the actual principal balance of the retained portfolio is $791 billion or so, but the book value is $760 billion. The difference is in the loss reserve. Which is by definition available to cover near-term losses.
For rather obscure reasons, OFHEO made both agencies start reporting UPB of the retained portfolios on their monthly volume summaries. The consolidated financials actually show the net balance after reserves.
It's important because the Paulson plan has the retained portfolio UPB capping out at I think $850 billion at the end of next year. So they're already at $791 billion, they may not want to load up on $10 billion or so in delinquent MBS loans right now.
Another excellent post. I always find your mortgage industry explanations invaluable. The length is always perfect - you write just enough to spell things out as simply and accurately as possible. Thanks for doing this.
"That probably refers to instances where they're using the accrual instead of cash method."
All companies use accruals, except the smallest home businesses.
Fair Value: Forget about it. No one is going to seriously introducing new accounting treatment in the middle of this mess. It is also just another set of estimates. For anon who is crazy about it, fair value includes fair value of liabilities also. Like debt. Look for other financial institutions that are booking increases in capital by reducing the "fair value" of their debts. Find at least one and then report back.
Andreas: Don't bet on International standards being any better. The motivation was to put all financial entities on an equal footing. Banks, insurance companies, etc. And also get rid of arbitrary, legacy capital regulatory requirements. A lot of those rule of thumb, legacy requirements are why the entire system hasn't collapsed yet. I.e. Legacy financial institution wanted lower capital requirements to compete with less regulated competition.
"Andreas writes:
This reminds me of an esteemed Congressman I saw interviewed (CNN?) a month or so ago, making the claim that the GSE's have an advantage as they don't have to mark their portfolio to market "
Banks don't write their portfolio of held mortgages to market either. Among other things, that means their value would change with each 1/8% rise or decline in interest rates. They do have to accrue for expected losses, which is judgmental. It is also disclosed and easy to find and people can make up their own minds regarding how realistic they are. Among other things, that is the theoretical function of security analysts - superimposing their judgment on qualitative aspects of the balance sheet and earnings.
The financial statements of F&F are complicated. Now that they are pretty much out of contention as investments (equity and anything junior of the govt), why bother now?
As far as cost of capital, the "threat" of Freddie to quit buying additional mortgages for their held portfolio, which would have amounted to a reduction of billions a quarter would have been very smart consider the cost of capital associated with holding them. In my opinion, one of the reasons they got taken over.
There are far bigger elephants in the room regarding F&F's accounting and risks then the TIMING of moving impaired loans from one bucket to another.
The problem with the stupid reporting is that stuff either goes up or down, AND the reporters want/need a reason for everything. They have a 50% of being right for the wrong reasons.
Just because their reasoning is faulty doesn't mean they don't have it right, in spite of themselves.
Tanta writes:
"maybe it's that 'double-entry bookkeeping' scam we were talking about yesterday. Everybody knows double-entry means 'lies.'"
Hold on just a minute. I strongly dispute the contention that "everybody knows double-entry means 'lies.'" Double entry bookkeeping is merely a fancy way of saying debits must equal credits. This can be confirmed by anyone with even the most rudimentary knowledge of accounting.
By the way, accounting problems can be roughly sorted into buckets.
Worldcom = They didn't move stuff into different buckets. They just made things up. And wrecked their competition, like ATT, in the process as a result of their lying.
Enron = They went to the trouble of inventing profits using something similar to real accounting. However, very little in real profit.
Almost everyone else -- wrong side of the line. didn't know what they were doing. other other types of things.
"Ed writes:
Thanks, what a mess this is...even people who are not unintelligent have to really think about it, where does that leave the herd?"
Exactly. Take me for instance. I somehow strayed onto this page looking for some cud when I ran into this post. I'm going to have heartburn for a week.
Tanta...thank you for writing a detailed overview. It helps decipher what is real life & how it works and gives a more informed basis for investment decisions.
Also on the plus side, it allows me to look amazed at ISIL (idiot-son-in-law) when he sums up the thing with "it's just the crooks getting rich" with not a whit of understanding, or any desire to have it.
Tanta, DO YOU THINK THAT BECAUSE EVERY ONE HAS A SHORT ATTENTION SPAN AND CANNOT READ A LONG POST ON ACCOUNTING MEANS THAT WE HAVE BEEN COMPLETELY DUMBED DOWN AS A NATION?
Should we try to summarize it in 2 words or less for them? That seems to be the median attention span for the American voting public.
Tanta: I know I'm late to the party, as usual, but kudos to you for explaining, and re-explaining this. I do, however, think the main question on everyone's mind here is about the losses associated with these "assets"--i.e. the actual houses underlying the loans on Freddie's books. You say, "Put-backs of loans generally only happen when the loans have defaulted, so the put-back then counts as a "recovery" of losses already realized, I believe." So I'm thinking we ought to be able to see those losses realized in a clear way and on a standard schedule. You're saying this is a footnote at best, in the "credit risk" section. As someone out here looking at some of these actual houses repurchased (at foreclosure auction) in recent months and weeks by Freddie, I'm thinking that this "credit risk" part is going to maybe be a much bigger deal than we've been generally led to believe. I'm seeing loan amounts in the $300,000 range on houses worth, at best, $100,000. Bunches of them. The loan originator(s) and seller/servicers (to which the loans can be put back)? Already toast. The mortgage broker(s)? Probably criminal, certainly judgment proof. So, how many of these assets are there (somewhere) on Freddies books? And given that each asset is good for $100k (tops), and the attendant unpaid debt on each is $300,000, then arent we really talking about losses (in these cases) of $200,000 a pop? And if so, then isnt the question of just when those losses are recognized, in a fully-accounted-for-sense, pretty important? This might be off-topic, and I know it's late. But it does seem to be the crux of the biscuit.
If we disregard all the on-off balance sheet, accounting requirement regularities, etc.
Freddie and Fannie have guaranteed something like 5 trillion in payments.
You then therefore come up with some sort of estimate of what % of those loans will go in default, and let's get aggressive and say 10%, and they are on the hook for 500 billion.
then lets say, of those loans, after the cost of forclosure and the net depreciation of the underlying asset, these companies can sell off the house for half of the loan value, so therefore the loan losses would be 250Billion.
This is consistent (250B) with the upper limit of losses I've seen projected out there.
If you wanted to estimate a more accurate assumption, any of the variables listed above could be changed, but the same calculation of the cost to the taxpayer would be made regardless.
Tanta- All I was pointing out earlier is when you stated that the asset sold "is on the balance sheet ...here" When in actuality, it's only the 1.25% guarantee fee they collect which was on the balance sheet.
I realize through your post and subsequent posting you understand the concepts, just that that the wording on your original post might be confusing to some.
The only way/reason I looked into it is I keep hearing that Fannie and Freddie "half half the mortgages in the country, which is 5 trillion" Then I look on their balance sheets and see 1 trillion worth of total assets and total liabilities, and I was puzzled.
A second issue in the article that is somewhat deceiving is the comment on the deffered tax assets:
"Freddie Mac and Fannie Mae have also inflated their financial positions by relying on deferred-tax assets credits accumulated over the years that can be used to offset future profits."
This is obviously another misunderstanding of what these deffered tax assets are. A deferred tax asset is created when an instrument that is not held in a mark to market through income(ie trading) portfolio but rather a portfolio that is market to market on the balance sheet but not through the income statement (ie available for sale). Since gains or (in this case) losses have no tax impact (as they don't accrue to income) the loss flows through to an other comprehensive income line which offsets equity. Before this happens, the tax impact is recorded as a deffered tax asset so that the full loss does not flow through income.
Net net, it reflects the tax benefit for a loss the company hasn't yet taken and may/probably will never take. If this tax item did not exist, it would also not impact regulatory capital as other comprehensive income (the liability that the tax asset offsets) is added back to book equity to come up with capital.
if you've ever met jim lockhart, ed demarco, phillip swangle or any of the rest of the cabal that pulled off this hostile takeover, you'd know that they are financial illiterates.
they were chosen for ideology, not competence. that's why they can't even properly explain to the reporters to whom they leak stuff the actual capital and economic implications of the accounting policies the GSE's chose.
Andrew: That's quick, dirty, certainly oversimplified yet somehow elegant. I want to believe it--so I know it must be wrong. Andy: what you write is so twisted and senseless that I want to scream and then throttle the first tax lawyer I can find. So I know it must be true.
Um Tanta? Is there anything below the fold? Thanks.
It seems to always take a few minutes to appear.
Can't you give us a twenty five words or less summary? I would rather eat ground glass than read a long post on accounting trivia. Just saying.
It seems to always take a few minutes to appear.
It always takes at least several seconds and sometimes a minute or two, depending on how fast Blogger is running. If you see one of these posts without a link at the bottom, just wait a bit and refresh and you will see it eventually.
Mad Dog Denninger worth a read today:
"While you're at it, consider the non-agency (private label) mortgage market. Think west-coast (in particular) lenders with big Option ARM exposure. These guys are toast. WaMu, for example is trading today as if an FDIC takeover is imminent; the October $2.50 PUTs have traded over 10,000 contracts (as of 9:25 CT) against an O/I of 27,813 this morning - clearly, someone thinks they're a big fat zero between now and October expiration and is willing to shoot a decent wad at that bet."
The Market Ticker
I would rather eat ground glass than read a long post on accounting trivia.
I would rather eat ground glass than try to summarize this post in 25 words. Just sayin'.
But maybe someone who feels more deeply your pain will summarize it in the comments for you.
There should be two distinct entities within Freddie and Fannie. One for the retained, and one for the guaranteed.
Throwing them together just complicates things needlessly.
Do the same capital requirements apply to retained and guaranteed? If not...why the heck not? Are they risk-based requirements?
Do individual loans move from the guaranteed to the retained in the event of default?
Is there a metric for tracking this activity?
LEh is down $4 dollars. Guess KDB decided bridge too far. Fuld is toast.
Looks like Jamie Fuld is getting a severe case of vertigo.
Thanks Tanta! That made sense to me...and while complex, not nearly as crazy making as, for instance, someone describing a derivative to me, or the alternative minimum tax...
O.K.,O.K. I will try to read it but may have some kind of breakdown before I reach the end.
Do individual loans move from the guaranteed to the retained in the event of default?
I obviously am incapable of explaining this to you, since I just wrote a big honkin' post on the subject and it didn't do anything for you. Maybe someone else can try explaining it again?
Whoa, the effect of Hank "Pimp my market" Paulson's plan lasted ... all of 24 hours! Way to go. Next bailout please.
as to not offend tanta, how bout a OpenT, leh is going down , Hard today.
Sorry Tanta, yeah should have suggested an open thread before posting about LEH.
Tanta writes:
Do individual loans move from the guaranteed to the retained in the event of default?
I obviously am incapable of explaining this to you, since I just wrote a big honkin' post on the subject and it didn't do anything for you. Maybe someone else can try explaining it again?
You actually expect me to read all of that?
WM is down 17%. Got cash, better run.
Have we seen this movie before?
Wonderin though, could this start a bank run.......
But maybe someone who feels more deeply your pain will summarize it in the comments for you.
Tanta | Homepage | 09.09.08 - 11:09 am | #
Tanta, could you please make it into a haiku? Thanks.
Rather than gnash teeth over the niceties of legal/acceptable or descriptive terms versus terms of art I'm more concerned with whether holding for up to two years is good policy. In a period of declining values both the holding costs and depreciation are dramatically increasing the lost opportunity penalty. Just because they can hold up to two years doesn't mean they should. I also see a slippery slope in that they might find themselves end loading their workload. That would probably result in bulk disposals at even lower prices.
Re: quasi-religious faith in some quarters,
I have unlimited faith in this and see no reason to question it: Sarah Pali
"What follows is my best effort to discover what the hell these people are talking about. I must disclose to you all that I am really just making an educated guess here."
and expecting them to use their brain as you would..
but that's asking alot!
You may be correct on this issue Tanta but I still think your boyfriend Dick has got some 'splainin' to do.
Tanta...I seem to recall a previous post (or coment maybe) that you made about them being able to send the loans back to the originators...in case of some structural defect or fraud on the part of the loan...am I making that up? If they can send them back, can they do that from both portfolios? I seem to recall further that there was a specific amount of time that they had to do this...and THAT is what I remember being 2 years...
This may be coming to a head. Lehman is "too big to fail" but the fed can't politically bail them out.
So do the twin towers have to set aside any capital for the guaranteed portfolio?
That's really what I wanna know.
Cause after RTFA, I still don't know.
If they don't I may have to head over to Lefty's post-haste.
Hanging - I suggest you get offline and turn on CNBC. It's better suited to your attention span.
The banks had research pieces out on this while ago, and none that I read were as lucid and straightforward as this piece. Thanks Tanta!
(Though in the end, I think they reached similar conclusions -- that this is doesn't change the nature of the business, it's simply a capital preserving move.)
IIRC - based on an interview I watched on Bllomberg TV some time agao- the answer is yes, but its a ridiculously small amount. Just a few bps.
Tanta --
Thank you for the write-up.
OK, so this is all about Fannie's/Freddie's decision whether to exercise their repurchase option.
If they make that decision based on what will be cheapest in the long run, that is one thing.
But if they make the decision based on the desire to defer losses, that is quite another.
The accusation is that they have been leaving loans in the guaranteed portfolio which they know damn well are going to continue to be non-performing. So they continue to make payments to the bondholders, even though they will eventually have to repurchase those loans anyway... And they are doing this solely to avoid recognizing losses today. (Again, this is the accusation.)
Now, the IHT article you linked says:
The company had made decisions that, while not necessarily in violation of accounting rules, had the effect of overstating the companies's capital resources and financial stability.
Indeed, one person briefed on the company's finances said Freddie Mac had made accounting decisions that pushed losses into the future and postponed a capital shortfall until the fourth quarter of this year, which would not need to be disclosed until early 2009.
Is this not a reasonable layman's summary of the accusation?
This post is about financial accounting matters. If you are one of those people who drove us insane in the comments to yesterday's post on "assets" versus "liabilities" by arguing that "assets" are "really" "liabilities" because you, like Humpty Dumpty, are The Master, then you will find this post unsatisfactory.
Thank you. Most excellent Lewis Carol reference.
I am, however, willing to sell my liabilities to several people on yesterday's thread, at bargain prices. Just sayin'.
Nice post Tanta. Quite clear. They don't take losses for two years...
DUCKS!
My guess the real story is in the liabilities side. All that short term paper needs to role fairly frequently with lots of volume. IIRC, they had a big rollover at the end of September.
My guess only, can't find anything, is that the cost of rolling was too high. I'm trying to find something, haven't yet, but that's the only thing that makes sense here.
Cheers,
The only "two years" rule that matters is that the only GSE paper that's even arguably backed by the bailout is the two year maturity, and even this is implicit, based on the minimum time period expected of the conservatorship. Paulsen did not use the magic words "full faith and credit" . This is not Treasury paper, quite aside from the convexity issue, which is arguably worse given that the explicit aim of the program is to lower mortgage interest rates. The limits of this alleged guarantee seem to have been overlooked by many but not all commentators. See WSJ: HEARD ON THE STREET, "Future Shock for 'Frannie' Debt" By DAVID REILLY
September 9, 2008; Page C14
I'm more concerned with whether holding for up to two years is good policy.
The problem here is that your unwillingness to deal with the technical details leads you to use the term "holding" here in a way that makes NO SENSE.
Look, whether the loan is in the retained portfolio or the MBS, someone is trying to bring it current (collect) or work it out or foreclose. These things take varying amounts of time. Foreclosure rarely takes more than two years, but in totally weird and usual circs it can.
Nobody can refuse to "hold" the loan while these actions are taking place.
The point is merely that the MBS can "hold" the loan during the default servicing process only for up to two years. After that, Freddie has to "hold" it. Somebody has to "hold" it until it is reinstated or liquidated or whatever.
"Do individual loans move from the guaranteed to the retained in the event of default?"
it was an easy reading, thank you tantachka.
...we purchase loans from PCs when the loans have been 120 days delinquent and...(b) foreclosure sales occur...
How can they buy a loan after a foreclosure sale? Doesn't foreclosure wipe out the loan?
...(c) the loans have been delinquent for 24 months...
The loan has been delinquent for 120 days AND 24 months? Doesn't make sense.
As to the rumored change extending to two years the period before bad loans have to be reported, I understood that to mean the retained portfolio. But if Tanta says it isn't so, I'm a believer.
We should create a debunking site, a la Snopes, for Tanta's best work. Name? "Dopes" comes to mind, but one of our posters might sue for infringement.
Apologies to tanta. why is the 2 yr situation any different from what Wells annoucned in its 2Q press release re delaying delinquencies to 180 days (I recall)? Merely a way to manage provisioning (smoothing function). Thuis also comes on heels of reports bacnks are simply delaying foreclosure. Also GSEs wrap has to buy back the defaulting mortgages in the pools and cure them. why would it be a surprise that they are delaying the repurchase as part of their wrap function? One would think in theory thsat the securities themsevels (pas throughs) would reflect the value of the NPL drag.
that was apologies on LEH
FYI posts from CRs vault which may or may not matter:
The Regulatory Reform Act also provides the Secretary of the Treasury with temporary authority, until December 31, 2009, to purchase any obligations and other securities Fannie Mae and Freddie Mac issue under certain circumstances. We intend to invest exclusively in Agency RMBS. The Regulatory Reform Act may lead to market uncertainty and the actual or perceived impairment in the credit quality of securities issued by Fannie Mae or Freddie Mac. This may increase the risk of loss on investments in Fannie Mae and/or Freddie Mac-issued securities. The Regulatory Reform Act could adversely affect the availability and pricing of Agency RMBS and, therefore, adversely affect our business. Welsh rarebit | Homepage | 09.06.08
A perfect example is SOP 03-3. Historically when mortgage underlying one of Freddie's MBS securities defauted Freddie would immediately buy it out of the trust, foreclose on it, and immediately book the loss. However, starting in Q1 2008 Freddie began to initially leave non-performing loans in the trusts hoping that they would cure. In effect, by delaying its purchase of non-performing mortgages out of the trusts Freddie began to delay its recognition of losses on the non-performing mortgages.GSE Guru | 09.06.08 - 7:25 pm
Fannie's Liquid Investment Portfolio (LIP) and Freddie's Liquidity and Contingency (L&C) portfolio are currently providing them with all the liquidity they need. Bear Stearns collapsed because it obtained most of its liquidity through overnight REPOs. The GSE's LIP and L&C portfolios currently do not have any REPOs at all! Unlike the major Wall Street firms Fannie and Freddie are in the business of holding securities until maturity. That means that as long as the mortgage delinquency rates do not get too out of line the value of the MBS securities they hold or guarantee should be irrelevant.
Anonymous | 09.06.08 - 10:21 pm
The stock market is a pull-toy for the hedge funds. Too bad most of Americans have their economic futures tied up there, in one way or another...
If they can send them back, can they do that from both portfolios? I seem to recall further that there was a specific amount of time that they had to do this...and THAT is what I remember being 2 years...
If the loan is defective (not just delinquent, but involving misrep) it can be put back to the seller/servicer from retained or MBS. And there is NO statute of limitations on that: the loan can be put back whenever the defect is identified.
It is rare that a 10-year old loan would be put back, because it's rare that a 10-year old loan that had performed for 10 years went bad and resulted in a loss. And defects are usually not found until the loan does go bad, and the GSE does a post-default quality control review. However, theoretically it can happen that way, because there is no time limit on the repurchase warranty.
Thanks Tanta!
Tanta, I am sure that you are right in your analysis. The principle is the same as a mortgage banker who finds himself obligated to repurchase a delinquent or otherwise defective loan from an investor opting instead to agree to sign an agreement indemnifying the investor against any losses incurred on the loan. Part of this indemnification would be advancing payments to reimburse the investor (servicer) for his advances on the MBS. The interest cost of the MBS advances is the coupon rate on the mortgage. Since the rate on the mortgage is fixed, one could easily have a defective low rate loan, say 5.75%, which becomes delinquent during a credit crunch where the mortgage banker's warehouse line cost could be considerably higher. For small mortgage bankers that could easily be the case, since warehouse lending agreements often call for higher rates and a much lower percentage of principal advance on repurchased loans.
On the other hand, the mortgage could have been originated during a time of relatively higher rates and if the time for repurchase comes after a Fed easing such as we have just experienced why would Freddie want to advance interest on a 6.75% loan when it had borrowing costs of 2.00%? Of course, this presumes that Freddie was capable of issuing more cheap "agency debt" and it would seem that recently that capability was called into question.
How can they buy a loan after a foreclosure sale? Doesn't foreclosure wipe out the loan?
All that really means is that the MBS never owns REO. By no later than the FC sale date, the MBS investor gets principal back.
The loan has been delinquent for 120 days AND 24 months? Doesn't make sense.
A loan could be 120 days delinquent for two years. That means that the borrower missed four payments, then started making payments again, but never caught up. That cannot continue for more than two years. (And it is never allowed to continue that long in the vast majority of cases.)
A few more FYI's from The CR vault:
Mortgage-Related Security Cash Flows Because losses on sold loans are absorbed by the Enterprises directly and are not passed through to security holders, no additional credit losses are reflected in cash flows calculated for an Enterprises own mortgage-backed securities (MBSs) held as investments. Cash flows for single-class MBSs issued by an Enterprise and held as investments consist only of principal and interest payments. www.ofheo.gov/media/archive/docs/regs/ RBCFinal.pdf capital-restoration plans | 09.07.08
Fannie Mae has also created an accounting policy related to its MBS pool activities, which appears to have as a primary purpose the avoidance of consolidation under FASB Interpretation No. 46, "Consolidation of Variable Interest Entities," also known as FIN 46. OFHEO believes that in this area, the Enterprise has not complied with FIN 46. As part of its policy, Fannie Mae transferred securities between SFAS 115 categories, yet did not provide the required documentation to describe the rationale for the transfers. http://www.ofheo.gov/NewsRoom.as...e=Detail& ID=216
I also see a slippery slope in that they might find themselves end loading their workload.
You lost me with this one, Rob Dawg. Nowhere do they say or imply that they're going to instruct their servicers not to take any action on these loans for two years.
Madam, go here:
Many entities are in the process of implementing FASBs recently issued guidance on fair value, SFAS 157, Fair Value Measurements, and SFAS 159, The Fair Value Option for Financial Assets and Financial LiabilitiesIncluding an Amendment of FASB Statement No. 115. Proper implementation can require CPAs to exercise professional judgment. This professional judgment is being put to the test in the measurement of fair value for mortgage-backed securities (MBS) because of significant declines in values and reduced levels of trades resulting from the recent credit market crises. This article illustrates the application of this professional judgment to MBSs as well as a discussion on other-than-temporary impairments of MBSs under SFAS 115, Accounting for Certain Investments in Debt and Equity Securities. In addition to providing a primer on the accounting treatment of MBSs, the discussion of SFAS 157 and SFAS 159 is expanded to help CPAs apply these standards in other situations
Mortgage-Backed Securities and Fair-Value Accounting
Worth a peek there!
Thank you for sharing your knowedge, Tanta.
An important exception exists for available-for-sale and held-to-maturity securities that have experienced an other-than-temporary decline in value. The exception, discussed below, requires that the unrealized loss be reported in the determination of net income rather than as a component of other comprehensive income.
The 2 years is simply the end of the option period. During the two years, Freddie can opt to either advance the payments to the investor at the mortgage MBS rate or buy the loan principle if Freddie's cost of funds is cheaper. They are responsible for the foreclosure losses regardless. The only issues are capital cost and cash flow. Until recently cash flow for GSE's had never been a problem.
The issue with Gretchen and other similar reporters is that they are writing about things that neither they nor their readers have any understanding about and they consistently get it wrong. They appear to have an agenda and anything they hear is filtered according to that agenda.
Thanks for the short course on MBS accounting. I knew Gretchen had to get it wrong somehow, I just know how.
How is it that she's considered a star in the financial section of the New York Times? Doesn't anybody there read this blog?
Thank you, Tanta.
This post is a perfect example of why I no longer bother to read any articles in newspapers anymore, except possibly the headlines in order to be aware of what others will say or be thinking on the topic.
would rather eat ground glass than try to summarize this post in 25 words. Just sayin'.
But maybe someone who feels more deeply your pain will summarize it in the comments for you.
Tanta | Homepage | 09.09.08 - 11:09 am | #
In Texan slag we are screwed,
2. screwed
3. screwed
4. screwed
5. screwed
6. screwed
7. screwed
8. screwed
9. screwed
10. screwed
11. screwed
12. screwed
13. screwed
14. screwed
15. screwed
16. screwed
17. screwed
18. screwed
19. screwed
20. screwed
21. screwed
22. screwed
23. screwed
24. screwed
You know Tanta, I think there was a conspiracy to destroy these great organisations. Remember it started in '03 for Freddie and '04 for Fannie.
What was the big deal about > 11 billion in "accounting errors"?
P.S. In '03 and '04 it took real talent to screw up in real estate.
P.P.S. If my memory serves me well, Enron's "accounting error" was ~1/2 billion. I don't have to remaind you what happened to Enron's boyz and what happened to F&F's boyz.
"What shows up on Freddie's balance sheet is the "guarantee asset," which is the fair value of the guarantee fees received, and the "guarantee obligation" (over on the liability side) which reflects the fair value of the projected credit losses."
Haven't been following closely, but are you saying there was no regulatory capital requirement for the guarantee book?
Disclosure requirements may discourage entities from violating the intent of the fair-value election. An entity must disclose its reasons for electing the fair-value option. If a partial election is made, an entity must disclose the reasons for the partial election.
If it is probable that the investor will be unable to collect all amounts due according to the contractual terms of a debt security not impaired at acquisition, an other-than-temporary impairment shall be considered to have occurred.
(PCAOB) emphasizes the importance of determining an entitys principal market when pricing services are used for fair-value measurements The alert states that a pricing service might provide an amount for which a large financial institution could sell the financial instrument. However, a company that owns that financial instrument might not be able to transact in the same market as a large financial institution. If the price available to a large financial institution would not be available to the company, then that price may not be an appropriate measure of fair value under SFAS 157.
The use of Level 3 inputs is necessary when observable inputs are not available. For MBSs, this could happen when very little or no market activity is observed for identical or similar assets at the measurement date. In using Level 3 inputs, SFAS 157 (para. 30) indicates that the fair-value measurement objective remains the same, that is, an exit price from the perspective of a market participant that holds the asset. Assumptions in a model used to measure fair value should take into account assumptions that market participants would make in pricing an asset if the information can be obtained without undue cost and effort.
The current environment of foreclosures, defaults, and uncertain real estate values suggests increased risk and uncertainty of future cash flows for investors holding MBSs. Accordingly, measuring risk for some MBSs may have become more difficult.
"Tell it to the Marines."
No. 4 years was enough. We would rather break stuff than listen...
Good explanation BTW.
Chris
As noted, fair-value measurements based on the use of unobservable inputs (Level 3) requires the exercise of significant professional judgment. Consistent with the subjectivity of these measurements, extensive disclosures are required by SFAS 157 (see paras. 3235). Among other disclosures, entities must disclose the levels (1, 2, or 3) in which inputs to fair-value measurements fall.
Re: Some of our financial instruments, such as our trading and available-for-sale (AFS) securities and our derivatives, are measured at fair value on a recurring basis in periods subsequent to initial recognition. We measure some of our other financial instruments at fair value on a non-recurring basis in periods subsequent to initial recognition, such as held-for-sale mortgage loans. Table 1 presents, by balance sheet category, the amount of financial assets carried in our condensed consolidated balance sheets at fair value on a recurring basis and classified as level 3 as of June 30, 2008. We also identify the types of financial instruments within each asset category that are based on level 3 measurements and describe the valuation techniques used for determining the fair value of these financial instruments. The availability of observable market inputs to measure fair value varies based on changes in market conditions, such as liquidity. As a result, we expect the financial instruments carried at fair value on a recurring basis and classified as level 3 to vary each period.
FEDERAL NATIONAL MORTGAGE ASSOCIATION FANNIE MAE - 10-Q - 20080808 - MANAGEMENT_ANALYSIS
Excellent summary. I'm new here, and would appreciate a preliminary clarification -- is Tanta's blunt persona part of her schtick? More to the point, while everything in Tanta's post makes sense, why is Nemo's comment above not still relevant?
Most important, an entity can make the election on the adoption of SFAS 159 for financial assets and liabilities owned or issued before that date (e.g., on January 1, 2008, for calendar yearend companies). Additionally, the election can be made on a security when the security is purchased. Paragraph 9 of SFAS 159 includes a complete list of triggers allowing for an election. Once SFAS 159 is effective for an entity, the effect of the first remeasurement to fair value will be reported as a cumulative-effect adjustment to the opening balance of retained earnings (para. 25).
The other day Tanta asked if someone had a checkbook, which was an allusion to the simplicity of accounting theory.
Are we to assume that we can manipulate checkbook ledger accounting for two years?
jkr0sc0 writes:
Excellent summary. I'm new here, and would appreciate a preliminary clarification -- is Tanta's blunt persona part of her schtick? More to the point, while everything in Tanta's post makes sense, why is Nemo's comment above not still relevant?
jkr0sc0 | 09.09.08 - 12:17 pm | #
You are asking for it...
LEH down 40% on an inability to balance the checkbook!
If LEH can't do simple math, maybe Paulson can?
why is Nemo's comment above not still relevant?
I may not understand Nemo's comment. It seems to involve the question of whether Morgenson and Duhigg accurately described an inaccurate accusation or not. I, personally, simply don't care about the answer to that question.
The "accusation" is that Freddie's repurchase policy has the sole purpose of delaying recognition of losses. I just wrote a long post suggesting that the purpose is more likely to reduce funding costs and hence reduce the ultimate losses to Freddie. I don't care how faithfully these reporters stenographed the accusation. Since they didn't follow it up with trying to find out if it's true or not, they don't get credit from me for simply repeating unsubstantiated claims.
ice work. Morgenson is irresponsible and unable to fully understand much of what she writes about.
I t is funny how thousands of banks are fucking up accounting and cant understand this basic accounting shit that average person should be able to understand, like Gretchen, Paulson, Fuld, Dodd, Tanta...
It appears Freddie is tightening the criteria under which they transact guaranteed mortgages into held mortgages. Morganson mistakenly thinks they are tightening the criteria under which Freddie is recognizing impairments.
The one plausible issue is that Freddie's internal procedures likely call for a more thorough review of valuation upon transacting mortgages to "Held". On the other hand, the inputs into valuation on "Held" portfolio would probably be more on the Level 3 end which are more judgemental and less empirical.
Would it be fair to say the the increased cost of capital for F&F had a side effect where MBS repurchases ( and the immediate write-down that does along with it ) were put off.
This could be perceived as an attempt to obscure the true cost of the market change by keeping them in the "guarantee portfolio" longer and delaying the inevitable write-down as market conditions deteriorate.
...I just wrote a long post suggesting that the purpose is more likely to reduce funding costs and hence reduce the ultimate losses to Freddie.
25 word summary
Haven't been following closely, but are you saying there was no regulatory capital requirement for the guarantee book?
No, I am not saying that.
They have to hold capital commensurate to their entire risk as an entire business. I'm not going to get into the exact capital requirements today.
It is simply that you calculate "risk" differently on the two sides of the business. In the retained portfolio you own the loans. You have funding costs that can--and apparently have, recently--exceeded the interest rate on the loans, which gives you a negative interest margin and blows a big hole in your income.
That is a risk you do not have in the MBS portfolio, because your investors have it for you. It's their principal outstanding, not yours. You just pass through interest at the coupon rate.
In both parts of the business you have credit risk.
Well, after yesterday, I now know the difference between liabilities and impaired assets. Ms. Wikipedia let me find some definitions and Mr. Google led me to a 10-k for Fannie. They have a description of their business activities that sounds similar to what Tanta just posted.
I only hope this conservatorship action works out better than some of the other dramatic actions the government has taken the last few years. Otherwise, see the remarks from FFDIC above.
ever underestimate the predictability of stupidity
thanks tanta.
if the millions of people who want to be informed citizens and read the nyt everyday continue to be subjected to this crap we are truly fucked.
you were too nice to Morgenson and Duhigg. virtually every word in the two paragraphs you posted were wrong.
This could be perceived as an attempt to obscure the true cost of the market change by keeping them in the "guarantee portfolio" longer and delaying the inevitable write-down as market conditions deteriorate.
Or, it could be perceived as finding the least expensive way to hold delinquent loans while servicing is running its course: invest your own (borrowed) capital, or let the bondholders continue to invest capital.
Now that we the peeps are the conservators of Freddie, which do you think we should make them do? Pay off bondholders faster than they have to and increase costs to us, or let the bondholders carry the loans until foreclosure?
How many times do I have to point out that the guarantee program takes losses, too? "Delaying" showing a loss in the retained portfolio just means that it continues to show up in the guaranteed portfolio (as an increase in the guarantee obligation).
You continue to rock. Thank you.
"I'm new here, and would appreciate a preliminary clarification -- is Tanta's blunt persona part of her schtick? "
IME, Tanta is impatient with sloppiness. She is particularly impatient with people who don't read what she wrote and then demonstrate it by asking questions that were just answered or making statements that were debunked in her post. She is actually quite patient and helpful with questions, even dumb questions, in other circumstances.
Thanks re capital clarification.
Do you have any comment on the tax asset criticism? This was also mentioned in a Bloomberg article.
Tax assets in my view are equivalent to deferred cash inflows (offsetting future taxes), albeit contingent on future profitability.
What I see in these criticisms is a tendency to associate capital with these specific assets, and an inability to get cash out of them when needed. But this is a false association in that it confuses the aspect of capital and liquidity. Write-offs and lossses don't necessarily generate a requirement for cash directly.
I'm also confused by the comment that financial institutions are not allowed to carry tax assets, which I've also seen in several places. If that's true, how do they square the balance sheet with the income statement when the after-tax cost of losses is less than the pre-tax cost?
"GSE" is sooo last week. Shouldn't we be leading the way by referring to them as the "GCE"s now?
Excellent post, btw.
"How many times do I have to point out that the guarantee program takes losses, too? "Delaying" showing a loss in the retained portfolio just means that it continues to show up in the guaranteed portfolio (as an increase in the guarantee obligation)."
Only if the inevitable future write-downs are properly documented. This is especially true with our current market where there is virtual certainty that the underlying asset is loosing equity every day ( the write-offs are only going to get worse ).
Don't get me wrong, I agree that not exercising their option makes the most sense today.
Tanta --
The "accusation" is that Freddie's repurchase policy has the sole purpose of delaying recognition of losses.
No, the accusation is not about the policy. The accusation is that F&F have lately been misusing their repurchase option specifically to delay recognition of losses, and thereby exaggerating their capital cushion.
Quoting from the IHT article again:
Indeed, one person briefed on the company's finances said Freddie Mac had made accounting decisions that pushed losses into the future and postponed a capital shortfall until the fourth quarter of this year, which would not need to be disclosed until early 2009.
IHT is citing a source who was "briefed on the company's finances", in the context of Morgan Stanley delivering its findings. I am not sure on what basis you say the accusation is "inaccurate", when it appears to be a simple and accurate layman's description of what Morgan Stanley (apparently) discovered when they reviewed the companies' finances.
Thanks for clarifying, Tanta. Look forward to reading you regularly (including maybe a possible future post on capital requirements?).
Straight from Freddie's 2007 Annual Report.(page 27-28)
Minimum Capital. The minimum capital standard requires us to hold an amount of core capital that is generally
equal to the sum of 2.50% of aggregate on-balance sheet assets and approximately 0.45% of the sum of outstanding
mortgage-related securities we guaranteed and other aggregate oÅ-balance sheet obligations.
There are other risk based capital requirements. It seems they look at minimum, critical, and risk-based and take the highest requirement.
All of them seem to be a tad bit thin.
There are a handful of people that read your posts like this and continue to tell you that you are right.
The rest of the world doesn't 100% understand what is happening but at least investors are smart enought to understand the common is basically worthless.
Mortgage rates have lower a little bit but in time they will be head higher.
Tanta, you've made a very good explanation for why a company would do what they did.
But the really interesting question is:
When did Freddie start leaving the mortgages in securities longer? Did it correspond to when their cost of capital went above mortgage interest rates (making it an economically sound decision), or was it when capital became constrained?
Only if the inevitable future write-downs are properly documented. This is especially true with our current market where there is virtual certainty that the underlying asset is loosing equity every day ( the write-offs are only going to get worse ).
You are still confused.
Here's a classic example of how this works: an MBS loan goes 90 days delinquent. The servicer starts FC. FC takes another six months. That means that the total time this loan has to be held by somebody while it is delinquent is 9 months.
Freddie has an option: it can buy the loan out of the pool at 90 days down, meaning that it holds the loan in the retained portfolio during the six months it takes to FC. Or, it can leave the loan in the MBS during the six months it takes to FC. Either way, recoveries in FC are not affected: the liquidation proceeds are what they are. This policy does not increase the time it takes to liquidate loans. Therefore this policy does not increase credit losses.
Tanta, I actually liked the post and could basically understand it.
Just to set the record straight about yesterday, I never said that assets = liabilities or any such nonsense. I merely said that there was no way JSP (ie the Treasury) was going to wind up WITH the assets, he will wind up with the liabilities only.
I realize that there are some quite good assets in there, JSP just won't be getting any, that's all.
When did Freddie start leaving the mortgages in securities longer?
From the 10-K thing I quoted in the post:
Through November 2007, our general practice was to purchase the mortgage loans out of PCs after the loans became 120 days delinquent. Effective December 2007, we no longer automatically purchase loans from PC pools once they become 120 days delinquent,
I'm also confused by the comment that financial institutions are not allowed to carry tax assets, which I've also seen in several places. If that's true, how do they square the balance sheet with the income statement when the after-tax cost of losses is less than the pre-tax cost?
If that is true (and I've never done accounting for financial institutions), then one would have an allowance on the balance sheet (and an expense on the P&L), that would exactly cancel out the asset. This is often done with startups that may never be profitable, or won't be profitable without significant new capital (and the changes in ownership that come with new capital). There should be a Note accompanying the financials explaining the treatment of deferred taxes.
Did it correspond to when their cost of capital went above mortgage interest rates (making it an economically sound decision), or was it when capital became constrained?
Anonymous | 09.09.08 - 1:03 pm | #
I'm a bit confused here--what's the difference between capital being constrained and capital being expensive?
This reminds me of an esteemed Congressman I saw interviewed (CNN?) a month or so ago, making the claim that the GSE's have an advantage as they don't have to mark their portfolio to market (and this wasn't meant as an accounting advantage, but this Congressmen thought this would mean less losses in fact), as they can hold their loans to maturity. Ignoring for the moment whether or not this is an "advantage", though clearly the GSE's have exploited this accounting loophole (under the theory, presumably, that there is no established market for mortgages and so they carry at book value, ignoring the fact that there is a huge market for mortgages and that in fact the market treat mortgages as more-or-less fungible, at least enough so that it would not be difficult to establish a market avlue for the GSE portfolios, but I digress), the fact is you can't hold a mortgage to maturity when it defaults and forecloses and the entire reason MBS's from private-label issuers are trading at a massive discount to par is the expectation of these defaults and foreclosures.
In other words the only "advantage" the GSEs have is they get to delay recognizing losses, not that they avoid them.
But these kinds of accounting rules (if I hear one more person tell me about how "transparent" our financial system is and how great US GAAP is, I think I'm going to explode
) is why relying on financial statements in the US market is absurd.
I haven't studied the new international accounting rules in detail but I sure hope they are less prone to fraud and manipulation than US GAAP.
I urge all of you to just pray harder and everything will be fine!
You have the general gist of what is occurring pretty spot on, however there is correction/distinction I would like to make.
You state "What you see on the balance shee is a guarantee asset"
Note that on the balance sheet, the only assets listed are the assets they actually own. (I.E. total assets = 800 billion)
These guarantee assets are actually held in a trust, that is completely off the balance sheet. (Although I do notice a guarantee asset,net on their balance sheet which I'm assuming is net of the liabilities, so I could be wrong here. However your general point about the different accounting rules between the classifications is correct regardless)
I.E. the reason they want to keep delinquent loans in the "guarantee" pool is because it's not on their balance sheet, thus they can hide the write downs.
This would be fine if they always re-bought the loans and put them on their balance sheet after 90 days delinquent, but delaying consoloditing these assets and marking them down is how they are smoothing or hiding their losses.
I was looking for the breakdown of these guarantee assets yesterday, and I noted on their 10-K it stated "Per the special SEC accounting rules, we do not have to disclose information about these off balance sheet guarantees. However, we have done so at www.fanniemae.comdisclosure"
Then I went to that website, and it's like a maze, and found nothing useful.
Furthermore, in thinking about this "guarantee asset"
the only way this could be created is if they spent cash on this, and then as they receive cash back they recognize revenue, which doesn't make sense from my understanding of the ways guarantees work.
Or that when they created this asset they recognized revenue, and then decrease the asset when they receive the cash.
So this 11B in guarantee fee "assets" could very very easily be a complete fabrication of revenue earned in the past. If they have lax standards on writing these "Assets" down, the overstatement of their earnings could be staggering, to the degree that they've actually lost on these guarantee payments (Since it's net of liabilities, if it's an actual liability, the earnings overstatement could be tremendous)
So, in summary:
--2 years is the limit on how long Fannie/Freddie can delay repurchasing nonperforming loans. This doesn't mean they won't repurchase any before 2 years, it's just their outside limit.
--All mortgages they purchase are technically assets (even if they are nonperforming ones), while the MBSs they issue to investors("PCs" in GSE lingo) are liabilities.
This is a bit longer than 25 words, but I essentially conveyed the same information as Tanta's 3000 word essay in fairly short order, did I not?
There now, that wasn't so hard, was it? (ducks)
Whyizzit we(U.S.) only take over the loser banks after they hit bottom? I say we take over a few good banks to offset the bad banks and take back the 28 mil we just paid to frannie's ceo's. Where's the accountability?
I urge all of you to just pray harder and everything will be fine!
This must be God's accounting fraud plan?
These guarantee assets are actually held in a trust, that is completely off the balance sheet. (Although I do notice a guarantee asset,net on their balance sheet which I'm assuming is net of the liabilities, so I could be wrong here
Look, I intentionlly used Freddie Mac's terms for these things so that if you wanted to you could find them on the balance sheet. If you look at the consolidated balance sheet from the last Q, under "assets" you find this line for "Guarantee asset, at fair value" at $11 billion. Pop down to "liabilities," and you see this line called "Guarantee obligation" at $14 billion. The difference between those two numbers means that credit loss-related costs on the guarantee portfolio are larger than the value of the guarantee fees collected. Yuk. I never said they weren't taking losses, remember.
Notice that under liabilities you also see this "Reserve for guarantee losses on Participation Certificates." That would be a function of the fact that the guarantee obligation is larger than the guarantee asset.
Nemo:
Putting aside the normative question of whether entities should legally minimize regulatory capital requirements (especially if they're publicly announcing the way they will be doing so), I'd second Tanta's core critique -- that the authors of the IHT article appear to have incompletely understood and investigated their own accusation. To me (and seemingly to Tanta), the clincher is the reference in the article to the companies injecting their own capital into pools of securities containing these loans -- as Tanta points out, such injections are the very thing F&F are trying to avoid.
If I understand Tanta's responses correctly, she's not closing the door on the possibility that F&F undertook the change in their repurchase policy due to accounting/capital requirement benefits, although she does appear to be asserting that economics, not accounting/capital requirements, was the driving factor. But she doesn't want to give credit to reporters who simply parrot accusations without adequate comprehension and/or follow up.
The two things still missing for me: (i) do increases in the guarantee obligation roughly track the write-down that would occur upon purchase of the loans by F&F and (ii) even if losses are roughly equivalent under either silo, do capital requirements ultimately differ (although I'd need a primer on capital requirements to even begin to parse the latter -- I don't think it's as simple as using the percentages 3 cents so helpfully posted)?
You have funding costs that can--and apparently have, recently--exceeded the interest rate on the loans, which gives you a negative interest margin and blows a big hole in your income.
what exactly do you mean by "funding costs"?
How about a "Pay to Pray" program?
No pay, no pray?
All mortgages they purchase are technically assets (even if they are nonperforming ones), while the MBSs they issue to investors("PCs" in GSE lingo) are liabilities.
Well, no.
The MBS loans are off-balance sheet. They are therefore neither assets nor liabilities on the balance sheet. They are assets of the securitization trust.
What is on Freddie's balance sheet is its financial guarantee of those MBS.
The accounting issues are a complete red herring -- these 'issues' have been well known, nothing new, and nothing shady. Ditto capital ratios.
After all, Lockhart presided over pledging the books clean, and also presided over lifting the caps and reducing the capital ratios! And just in the last few weeks, Lockhart, Bernanke, et al all said there were no capital adequecy problems at the GSEs.
And yet, now, suddenly, Paulson has to move quickly to place the GSEs into conservatorship -- and he places Lockhart and Lockhart-selected people in direct control of the GSEs! If these new accounting problems were are real issue, why continue with Lockhart?
Obviously, a red herring.
But what I really want to know, is why isn't this aspect being discussed in the (free) press?
WTant Hon,
What about indexed securities linked to these MBS guarantees? How does that play into your world of liabilities here?
what exactly do you mean by "funding costs"?
I meant what the Q calls "the expected cost of holding the nonperforming mortgage in our retained portfolio."
What about indexed securities linked to these MBS guarantees?
I'm afraid I don't understand your question.
Hi Tanta,
An off-topic question: I think I understand correctly that Fannie and Freddie invented mortgage backed securities -- and basically securitization itself. But we they also the founders of CDO style (multi-tranche) securitization or was that an innovation that they adopted from the private sector?
My very profound thanks for your willingness to share your expertise with us on this blog.
Excellent post. Understandable, and I learned some stuff. Thanks.
Schtick? Pffft. Hardly.
The MBS loans are off-balance sheet. They are therefore neither assets nor liabilities on the balance sheet. They are assets of the securitization trust.
What is on Freddie's balance sheet is its financial guarantee of those MBS.
I stand corrected --thanks for the assist! Is it any wonder the mainstream press or public gets confused about this stuff, when regular readers of this blog can't even get it 100% right?
Personally, I think complexity, opacity and obfuscation (ex: hiding liabilities "off-balance sheet") are a built-in feature of mortgage securitization. Banksters know that regular investors and taxpayers have no comprehension of or patience for this kind of stuff, and that's how they pull off these shenanigans and get away with it.
(i) do increases in the guarantee obligation roughly track the write-down that would occur upon purchase of the loans by F&F and (ii) even if losses are roughly equivalent under either silo, do capital requirements ultimately differ (although I'd need a primer on capital requirements to even begin to parse the latter -- I don't think it's as simple as using the percentages 3 cents so helpfully posted)?
Well, that is certainly where this all gets very difficult. Remember that any loan that is modified has to be bought out of the MBS when it is modified. That has not changed. So presumably they look at a net present value calcuation (losses in FC vs. loss if you modify, which automatically means you take it to the retained portfolio), and go with "less loss." I think that is probably what the reporters were trying to get at by saying "The companies have injected their own capital into pools of securities containing these loans, arguing that their new policies are helping more borrowers." I have no idea how you could do the relative calculation of cost/benefit if the loss calcs between the portfolio and the guarantor program were just totally incommensurate. But they are doing a lot of mods, so we have to conclude that in some cases they'd rather take the immediate write-down to the retained portfolio than to foreclose and let the MBS carry the loan until liquidation.
It is always difficult with the outfits formerly known as GSEs to know when this is an economic decision, when it's a capital/regulatory decision, and when it's that "mission" thingy. Nobody ever puts a line on the financials for "costs incurred in keeping Congress happy by preventing unnecessary foreclosures" or whatnot.
I don't assume that the Fs have ever done anything out of absolutely "pure" motives. I am just never quite sure what people mean when they talk about things being done "only for accounting purposes."
I'm afraid I don't understand your question.
Thank you Tanta for pointing out the inaccuracies and spin put out by the media, but is there a primer available somewhere for what went wrong at the GSE's for those lacking higher education, let alone an accounting background? I had tried reading that 10-K on Sunday;, and while today's post helps a great deal, I still have newbie type questions that I'd rather not ask if answered elsewhere.
I'm just asking for directions to elsewhere. thank you
What's the bottom line at the end of the day? That's really THE issue isn't it? What increase of non-performing mortgages are we talking about and what $ amount is now needed?
How MUCH capital would now be needed if they were to take the guarantys onto their investment portfolio, regardless of the relative expense - advantage or disadvantage to funding costs?
How much are the losses if they were to simply assume the issues NOW?
You do a great job by the way. My question is that whether you're financing $100bn at 5% or 6% the rate isn't as important as the amount...
Ed.
Nobody ever puts a line on the financials for "costs incurred in keeping Congress happy by preventing unnecessary foreclosures" or whatnot.
Nor for "jiggling the accounting to deceive investors and stave off regulatory action"
Does Fannie exchange indexed securities?
Re:hese transactions create what are called mortgage-backed securities (MBS). Fannie and Freddie only guarantee to the holders of MBS that the principal and interest on the mortgages in the trusts will be paid in full and on time. In other words, Fannie and Freddie take only the credit risk on these mortgages; the interest-rate risk--that interest rates will rise and the mortgages will become less valuable, or that interest rates will fall and the mortgages will be prepaid and disappear--is taken by the holders of the MBS. For their guarantee, Fannie and Freddie receive an annual fee as long as the pools remain in existence. In this segment of their business, Fannie and Freddie have guaranteed about $2.2 trillion in mortgages.[1]
In the other part of their business, Fannie and Freddie buy mortgages from originators and hold them in portfolio; they also repurchase for their portfolios some of the MBS that they have already issued. Because they are then the owners of these mortgages and MBS, they have assumed the credit risk, and because they have to borrow the funds to buy the mortgages and MBS, they assume the associated interest-rate risk. Of the two, interest-rate risk is far more significant, and Fannie and Freddie have to enter into large hedging transactions to mitigate it. They assume these substantial risks, and contract for hedging, because their portfolios of mortgages produce by far the greater part of their profits. This is true because, as GSEs, they are accorded favorable rates in the capital markets, and thus can arbitrage between their borrowing costs and the rate they will receive on the mortgages and MBS they hold.
AEI - Regulating Fannie Mae and Freddie Mac
Tanta,
I remember hearing a little while back that F or F was looking into requiring buy backs of mortgage paper sold to F and F where there essentially failure to comply with representations by the originator regarding conformity with F & F underwriting guidelines or early payment defaults.
Since you are very familiar with the way F and F does their accounting, my question is do these kind of contingent claims show up as any kind of significant entry anywhere on either of the F &F balance sheets, and if so, are they treated differently for portfolio vs. securitized loans?
Thanks.
I meant what the Q calls "the expected cost of holding the nonperforming mortgage in our retained portfolio."
i see. i think using the term "funding costs" is somewhat confusing, because it's not really about the cost of issuing agency debt (which is always substantially cheaper than the coupon on MBS) to fund the mortgage, and it doesn't really have anything to do with interest margin.
per Freddie:
Freddie Mac believes that the historical practice of purchasing loans from PC pools at 120 days does not reflect the pattern of recovery for most delinquent loans, which more often cure or prepay rather than result in foreclosure. Allowing the loans to remain in PC pools will provide a presentation of its financial results that better reflects Freddie Mac's expectations for future credit losses. Taking this action will also have the effect of reducing the company's capital costs. The expected reduction in capital costs will be partially offset by, but is expected to outweigh, greater expenses associated with delinquent loans.
Freddie Mac Announces Operational Changes for Purchasing Delinquent Loans From Mortgage PCs. - News Archive - Freddie Mac
You're right, I should have said "capital costs."
do these kind of contingent claims show up as any kind of significant entry anywhere on either of the F &F balance sheets
You won't find that on the balance sheets anywhere. Put-backs of loans generally only happen when the loans have defaulted, so the put-back then counts as a "recovery" of losses already realized, I believe. If the repurchase volume is significant, there should be a discussion of the matter in the footnotes in the "credit risk" section.
How MUCH capital would now be needed if they were to take the guarantys onto their investment portfolio, regardless of the relative expense - advantage or disadvantage to funding costs?
From the last 10-Q:
For the six months ended June 30,
2008, we purchased approximately $1.1 billion in unpaid principal balances of these loans with a fair value at
acquisition of $0.9 billion. Although the volume of these repurchases has decreased in the six months ended June 30, 2008, there was $12.5 billion unpaid principal balance of loans remaining in our PCs and Structured Securities as of
June 30, 2008 that were greater than 120 days past due for which we have not exercised our repurchase option.
So the answer, at the moment, seems to be about $12.5 billion less the fair value adjustment, whatever that is.
Your detailed posts are one of a kind wonderful, Tanta. They are a big help in trying to understand what's going on.
Thanks for that. I'm assuming you are referring to FNM...So a 10 fold increase in non-performing or delinquent mortgages....and added to the tax credit of $20.6bn takes a total of $33.1 bn from their regulatory capital with this calculation...accounting for 70% of the $47bn, leaving just $14bn to cover $800bn in their investment portfolio and $2trn some odd in guaranty book?
Tanta wrote "How MUCH capital would now be needed if they were to take the guarantys onto their investment portfolio, regardless of the relative expense - advantage or disadvantage to funding costs?
So the answer, at the moment, seems to be about $12.5 billion less the fair value adjustment, whatever that is."
No, no, no. The $12.5 billion is the face value of the loans not the capital required. Assuming an 8% capital requirement (generally applicable to banks) that would translate into $1 billion in capital required. However banks (don't know about Freddie or Fanny) usually have a capital charge associated with guaranties or off-balance sheet items. Therefore, the additional capital required to bring the $12.5 billion in loans on-balance sheet could be less than the $1 billion.
I am just never quite sure what people mean when they talk about things being done "only for accounting purposes."
That probably refers to instances where they're using the accrual instead of cash method.
No, no, no. The $12.5 billion is the face value of the loans not the capital required.
Sorry, I'm really getting tired.
$12.5 billion is what Freddie would have to pay the bondholders (par on UPB). $12.5 billion less the fair value adjustment is what would hit the retained portfolio (the difference being the write-down).
As you say, the capital reserve requirement would be less than that (if not 8%). But I thought Ed was asking a different question.
Ed, I'm talking strictly Freddie today. And remember that the retained portfolio balance is net of loan loss reserves. I'm getting too tired to look it up again, but I just looked this morning and I believe that the actual principal balance of the retained portfolio is $791 billion or so, but the book value is $760 billion. The difference is in the loss reserve. Which is by definition available to cover near-term losses.
For rather obscure reasons, OFHEO made both agencies start reporting UPB of the retained portfolios on their monthly volume summaries. The consolidated financials actually show the net balance after reserves.
It's important because the Paulson plan has the retained portfolio UPB capping out at I think $850 billion at the end of next year. So they're already at $791 billion, they may not want to load up on $10 billion or so in delinquent MBS loans right now.
That probably refers to instances where they're using the accrual instead of cash method.
Or maybe it's that "double-entry bookkeeping" scam we were talking about yesterday. Everybody knows double-entry means "lies."
Another excellent post. I always find your mortgage industry explanations invaluable. The length is always perfect - you write just enough to spell things out as simply and accurately as possible. Thanks for doing this.
Tanta, could you please make it into a haiku? Thanks.
Lemme see if I can:
Gretchen Morgenson
Thinks Freddie Mac is a bank
Just like Wells Fargo.
"That probably refers to instances where they're using the accrual instead of cash method."
All companies use accruals, except the smallest home businesses.
Fair Value: Forget about it. No one is going to seriously introducing new accounting treatment in the middle of this mess. It is also just another set of estimates. For anon who is crazy about it, fair value includes fair value of liabilities also. Like debt. Look for other financial institutions that are booking increases in capital by reducing the "fair value" of their debts. Find at least one and then report back.
Andreas: Don't bet on International standards being any better. The motivation was to put all financial entities on an equal footing. Banks, insurance companies, etc. And also get rid of arbitrary, legacy capital regulatory requirements. A lot of those rule of thumb, legacy requirements are why the entire system hasn't collapsed yet. I.e. Legacy financial institution wanted lower capital requirements to compete with less regulated competition.
"Andreas writes:
This reminds me of an esteemed Congressman I saw interviewed (CNN?) a month or so ago, making the claim that the GSE's have an advantage as they don't have to mark their portfolio to market "
Banks don't write their portfolio of held mortgages to market either. Among other things, that means their value would change with each 1/8% rise or decline in interest rates. They do have to accrue for expected losses, which is judgmental. It is also disclosed and easy to find and people can make up their own minds regarding how realistic they are. Among other things, that is the theoretical function of security analysts - superimposing their judgment on qualitative aspects of the balance sheet and earnings.
The financial statements of F&F are complicated. Now that they are pretty much out of contention as investments (equity and anything junior of the govt), why bother now?
As far as cost of capital, the "threat" of Freddie to quit buying additional mortgages for their held portfolio, which would have amounted to a reduction of billions a quarter would have been very smart consider the cost of capital associated with holding them. In my opinion, one of the reasons they got taken over.
There are far bigger elephants in the room regarding F&F's accounting and risks then the TIMING of moving impaired loans from one bucket to another.
The problem with the stupid reporting is that stuff either goes up or down, AND the reporters want/need a reason for everything. They have a 50% of being right for the wrong reasons.
Just because their reasoning is faulty doesn't mean they don't have it right, in spite of themselves.
Tanta.....
Good effort.
Tanta writes:
"maybe it's that 'double-entry bookkeeping' scam we were talking about yesterday. Everybody knows double-entry means 'lies.'"
Hold on just a minute. I strongly dispute the contention that "everybody knows double-entry means 'lies.'" Double entry bookkeeping is merely a fancy way of saying debits must equal credits. This can be confirmed by anyone with even the most rudimentary knowledge of accounting.
Thanks, what a mess this is...even people who are not unintelligent have to really think about it, where does that leave the herd?...
By the way, accounting problems can be roughly sorted into buckets.
Worldcom = They didn't move stuff into different buckets. They just made things up. And wrecked their competition, like ATT, in the process as a result of their lying.
Enron = They went to the trouble of inventing profits using something similar to real accounting. However, very little in real profit.
Almost everyone else -- wrong side of the line. didn't know what they were doing. other other types of things.
This can be confirmed by anyone with even the most rudimentary knowledge of accounting.
I take it you didn't see yesterday's comment thread on the "asset/liability" problem?
I was, you know, being sarcastic. I have been in favor of double-entry bookkeeping since the 15th century.
Double entry?? That means they have two sets of books..hah, I knew it.
I have been in favor of double-entry bookkeeping since the 15th century.
Tanta is 600 years old??
meh. Methuselah made it to 969.
But he was still never able to figure out why the GSE model made any sense in the first place.
Tanta is 600 years old??
No, that's when she read Pacioli.
"Ed writes:
Thanks, what a mess this is...even people who are not unintelligent have to really think about it, where does that leave the herd?"
Exactly. Take me for instance. I somehow strayed onto this page looking for some cud when I ran into this post. I'm going to have heartburn for a week.
Tanta...thank you for writing a detailed overview. It helps decipher what is real life & how it works and gives a more informed basis for investment decisions.
Also on the plus side, it allows me to look amazed at ISIL (idiot-son-in-law) when he sums up the thing with "it's just the crooks getting rich" with not a whit of understanding, or any desire to have it.
Thank you very much...
Tanta, DO YOU THINK THAT BECAUSE EVERY ONE HAS A SHORT ATTENTION SPAN AND CANNOT READ A LONG POST ON ACCOUNTING MEANS THAT WE HAVE BEEN COMPLETELY DUMBED DOWN AS A NATION?
Should we try to summarize it in 2 words or less for them? That seems to be the median attention span for the American voting public.
Tony
Tony: here ya go:
On Stupidity - Advice - The Chronicle of Higher Education
On Stupidity, Part 2 - Do Your Job Better - The Chronicle of Higher Education
Tanta: I know I'm late to the party, as usual, but kudos to you for explaining, and re-explaining this. I do, however, think the main question on everyone's mind here is about the losses associated with these "assets"--i.e. the actual houses underlying the loans on Freddie's books. You say, "Put-backs of loans generally only happen when the loans have defaulted, so the put-back then counts as a "recovery" of losses already realized, I believe." So I'm thinking we ought to be able to see those losses realized in a clear way and on a standard schedule. You're saying this is a footnote at best, in the "credit risk" section. As someone out here looking at some of these actual houses repurchased (at foreclosure auction) in recent months and weeks by Freddie, I'm thinking that this "credit risk" part is going to maybe be a much bigger deal than we've been generally led to believe. I'm seeing loan amounts in the $300,000 range on houses worth, at best, $100,000. Bunches of them. The loan originator(s) and seller/servicers (to which the loans can be put back)? Already toast. The mortgage broker(s)? Probably criminal, certainly judgment proof. So, how many of these assets are there (somewhere) on Freddies books? And given that each asset is good for $100k (tops), and the attendant unpaid debt on each is $300,000, then arent we really talking about losses (in these cases) of $200,000 a pop? And if so, then isnt the question of just when those losses are recognized, in a fully-accounted-for-sense, pretty important? This might be off-topic, and I know it's late. But it does seem to be the crux of the biscuit.
If we disregard all the on-off balance sheet, accounting requirement regularities, etc.
Freddie and Fannie have guaranteed something like 5 trillion in payments.
You then therefore come up with some sort of estimate of what % of those loans will go in default, and let's get aggressive and say 10%, and they are on the hook for 500 billion.
then lets say, of those loans, after the cost of forclosure and the net depreciation of the underlying asset, these companies can sell off the house for half of the loan value, so therefore the loan losses would be 250Billion.
This is consistent (250B) with the upper limit of losses I've seen projected out there.
If you wanted to estimate a more accurate assumption, any of the variables listed above could be changed, but the same calculation of the cost to the taxpayer would be made regardless.
Tanta- All I was pointing out earlier is when you stated that the asset sold "is on the balance sheet ...here" When in actuality, it's only the 1.25% guarantee fee they collect which was on the balance sheet.
I realize through your post and subsequent posting you understand the concepts, just that that the wording on your original post might be confusing to some.
The only way/reason I looked into it is I keep hearing that Fannie and Freddie "half half the mortgages in the country, which is 5 trillion" Then I look on their balance sheets and see 1 trillion worth of total assets and total liabilities, and I was puzzled.
A second issue in the article that is somewhat deceiving is the comment on the deffered tax assets:
"Freddie Mac and Fannie Mae have also inflated their financial positions by relying on deferred-tax assets credits accumulated over the years that can be used to offset future profits."
This is obviously another misunderstanding of what these deffered tax assets are. A deferred tax asset is created when an instrument that is not held in a mark to market through income(ie trading) portfolio but rather a portfolio that is market to market on the balance sheet but not through the income statement (ie available for sale). Since gains or (in this case) losses have no tax impact (as they don't accrue to income) the loss flows through to an other comprehensive income line which offsets equity. Before this happens, the tax impact is recorded as a deffered tax asset so that the full loss does not flow through income.
Net net, it reflects the tax benefit for a loss the company hasn't yet taken and may/probably will never take. If this tax item did not exist, it would also not impact regulatory capital as other comprehensive income (the liability that the tax asset offsets) is added back to book equity to come up with capital.
I think it is clearly poor reporting.
if you've ever met jim lockhart, ed demarco, phillip swangle or any of the rest of the cabal that pulled off this hostile takeover, you'd know that they are financial illiterates.
they were chosen for ideology, not competence. that's why they can't even properly explain to the reporters to whom they leak stuff the actual capital and economic implications of the accounting policies the GSE's chose.
Andrew: That's quick, dirty, certainly oversimplified yet somehow elegant. I want to believe it--so I know it must be wrong. Andy: what you write is so twisted and senseless that I want to scream and then throttle the first tax lawyer I can find. So I know it must be true.