Saturday, March 18, 2017

Tim Howard on FNMA's TDR-related reserves

Troubled debt restructurings is an “in-the-weeds” subset of the DTAs that I decided not to get into in this post, because it’s very technical. But since you asked….(and here again, for simplicity’s sake I’ll use Fannie’s figures, but Freddie has a similar issue).
One of the main ways FHFA and Treasury were able to run up Fannie and Freddie’s losses so much in the 2008-2011 period was through making mammoth additions to their loss reserves. At December 31, 2007, Fannie Mae had a single category of loss reserves totaling $3.4 billion. Four years later, at December 31, 2011, Fannie had five categories of loss reserves totaling $93.2 billion. Of this nearly $90 billion increase (every dollar of which resulted in the need to draw an equal amount of non-repayable senior preferred stock from Treasury), $63.2 billion were from actions that that required Fannie to designate the affected loans as Troubled Debt Restructurings (or TDRs). These TDR-related loss reserves fell into two categories: reserves on loans purchased from mortgage-backed securities pools ($16.3 billion), and reserves for loans the company modified ($46.9 billion).
If a loan is designated as a TDR, a company is required to use a very conservative “life of loan” accounting treatment that locks up the loss reserve for the life of the loan even if the loan returns to performing status. In a case of a loan modification– which a company does in the hopes that the loan WILL return to paying status– it has to write off immediately not only its estimate of future losses but also the present value of foregone interest payments. If the loan becomes current again, the written-off amount will be brought back into income– either as foreclosed property income (if the write down was an estimated loss) or as net interest income (if it was foregone interest)– only as the loan amortizes (i.e., extremely slowly) or in a lump sum when the loan pays off.
FHFA used TDR accounting very aggressively for Fannie during the 2008-2011 period. For example, the use of TDR accounting for the loans bought out of MBS pools ($16.3 billion as of December 31, 2011) was discretionary; GAAP did not require it–Fannie, that is FHFA, chose to do it. And the $46.9 billion in reserves for “individually impaired” modified loans was ballooned in two ways. First, Fannie (and Freddie) HAD to modify all loans that met the standards for Treasury’s Home Affordable Mortgage Program (HAMP), whereas banks could choose whether to modify HAMP-eligible loans or not. Second, Fannie’s write-downs on modified loans were far larger than other institutions’–an average of 27.5 percent of the loan balance, nearly triple the average 10.5 percent write-down of the largest bank modifier of single-family mortgages (Bank of America, whose loans were of much lower quality than Fannie’s).
Even today, nine years after the crisis and five years after Fannie’s peak loss-reserve total, Fannie still has $28.5 billion in TDR-related loss reserves ($21.9 billion in individually impaired single-family loans, and $6.6 billion in fair value losses on loans bought out of MBS pools). The vast majority of these reserves are on loans that are now performing, but Fannie can’t get at the reserves until the loans amortize or pay off.
FHFA (and Treasury) did a REALLY good job figuring out how to lock up for a very long period of time a substantial amount of what otherwise would have been Fannie Mae’s capital. 
It’s possible there are ways for Fannie and Freddie to unlock some of their TDR-related reserves, but I don’t know what those ways are. Perhaps their accountants do.

Saturday, February 25, 2017

Appeals Court Ruled Breach of Contract Claims are Ripe


http://www.fairholmefundsinc.com/Documents/PublicConferenceCall20161118.pdf
Fairholme Capital Management Public Conference Call
November 18, 2016



David Thompson: Yes, there are three standard remedies for a breach of contract.
One is expectancy damages, which puts us in the position that we would have been in if there had been no breach of contract. Two is reliance, which is to give us our out-of-pocket costs. The third is restitution. We’re entitled to present evidence of all three and pick the highest.
But, I want to focus on restitution, because I think that is really the concept that is the most relevant here, and it’s pretty simple. You look at the benefits that the breaching party received – and here the breaching party would be Fannie Mae and Freddie Mac – and the benefit they received was par value, $25 a share. From that, you would potentially subtract any benefits as they would probably argue for an offset of any dividends that the preferred shareholders received. Now as we know, two thirds of this float was issued in 2007 and 2008. So, for those series the offset from par value would be somewhere between zero and five dollars a share. Thus, we could be looking at damages of $20 a share if we are successful on our breach of contract claim and the court agrees with us about restitution. 

Saturday, February 18, 2017

Fannie Mae core capital - 2016 10-K page F-57

The following table displays our regulatory capital classification measures.
As of December 31, 2016 (Dollars in millions)

Core capital                                                                                             $ (111,836)
Statutory minimum capital requirement                                                    $ 24,351
 Deficit of core capital over statutory minimum capital requirement    $ (136,187)

The sum of (a) the stated value of our outstanding common stock (common stock less treasury stock); (b) the stated value of our outstanding non-cumulative perpetual preferred stock; (c) our paid-in capital; and (d) our retained earnings (accumulated deficit). Core capital does not include: (a) accumulated other comprehensive income or (b)  senior preferred stock. (2)   Generally, the sum of (a)   2.50%  of on-balance sheet assets, except those underlying Fannie Mae MBS held by third parties; (b)   0.45%  of the unpaid principal balance of outstanding Fannie Mae MBS held by third parties; and (c) up to  0.45%  of other off-balance sheet obligations,  which may be adjusted by the Director of FHFA under certain circumstances.

----------------------------------------------------------

If the senior preferred shares are recharacterized as a $116,149 loan that has just been repaid with interest, then the core capital increases to $4,313.  If some of the $23,465 allowance for loan losses can be recharacterized as core capital then even better.  With earnings at ~$10B per year, it would only take a couple of years to reach the statutory minimum capital requirement. However, it would take 7 or 8 years to reach Ackman's proposed $79B or Tim Howard's proposed $70B through retained earnings only.  Some dilution via a capital raise is probable because 7 or 8 years is too long for bureaucrats and politicians. 

Sunday, February 12, 2017

Fannie Mae valuation

Future common share price = future earnings available to common shareholders multiplied by P/E multiple divided by number of shares outstanding 

Tim Howard's FNMA estimate = $9.3B x 10 / 1.158 = $80 per share (undiluted)
Ackman's FNMA estimate = $10B x 14 / 5.79 = $24 per share (warrants exercised)
(Ackman's FMCC estimate = $5B  x 14 / 3.21 = $22 per share)

Ackman's plan is for FNMA to build up $79B (2.5% capital ratio) through retained earnings only.  Tim Howard's plan is for FNMA to build up $70B (2% single family; 3% multi-family; 2% for interest rate risk).  I do not believe that it will be politically viable to wait 8 years to build capital to these levels.  I believe that there will be an equity issuance to raise capital and that equity cannot be raised without cancelling all of the warrants.  It is hard to estimate the amount of share dilution that an equity issuance will cause, but I estimate that Tim Howard's estimate would be knocked down to $16 to $40 per share.

"There’s one final wrinkle I’ll just touch on here. As of September 30, 2016, Fannie had $22.5 billion in its loan loss allowance. That’s capital. But $21.3 billion of that is tied up in “life on loan” reserves on loans that were modified pursuant to Treasury’s mandatory Home Affordable Mortgage Program (HAMP). The vast majority of these HAMP reserves are against loans that are now performing, but the accounting rules require that the loss reserves be maintained until the loans either amortize or are paid off. If I were at Fannie—and searching for a way to reach full capitalization as quickly as we could—I would look very hard at ways to free up as much of that “regulatorily immobilized” capital as I could. Any amounts of loss reserves moved from the “individually impaired” to the “collectively reserved” category would reduce the amount of the equity capital the company would need to raise." - Tim Howard

Note: Proposed reductions in corporate tax rates would increase earnings, but DTA's would also have to be written down and capital would have to be raised (new corporate tax rate of 20% would cause a $15B write down of DTA's for FNMA & $8B for FMCC).

Sunday, December 25, 2016

114 North American Oil &Gas Bankruptcies Since start of 2015


http://www.haynesboone.com/~/media/files/attorney%20publications/2016/energy_bankruptcy_monitor/oil_patch_bankruptcy_20160106.ashx

Saturday, August 6, 2016

Thoughts on GSE's

1) Before the GSE shareholders can get anything the government is entitled to be paid back with interest.  ETA is 2017 year end.

2) Before the GSE's can be released from conservatorship they must have adequate capital.
   a) Someone has to make a decision and actually determine what is adequate capital.
      Historical ratio = 0.45% - near unanimous opinion that it needs to be more
      Ackman recommends 2.5% capital ratio or $123B
      Mulvaney bill called for 5% or $250B
      Corker-Warner and Johnson-Crapo called for 10% or $500B
      The UST has officially stated 10% too, but internal memo states 3-4% would be adequate.
   b) Where is this $125B-$500B going to come from?
      Ackman projects that just allowing the GSE's to retain earnings until the end of 2024 will build up capital to his recommended 2.5% ratio.  However, the FDIC precedent is to dilute the shareholders through a share issuance.  For example, if all of the warrants for 7.2 billion shares in both GSE's are cancelled (government is not entitled to anything after it has been paid back with interest), then the 2.5% capital ratio could be raised by selling 7.2 billion new shares in both GSE's at at price of $17 per share; a discount to the $23 per share as shown on slide 104 of Ackman's slide deck "It's Time to get off our Fannie".

However, if the government insists on a 10% capital ratio then a capital raise via a share issuance is a non-starter without higher g-fess (and therefore higher interest rates on mortgages).  Ackman's slide 104 shows that using a 14x earnings multiple and a 60 bps G-fee that the value of the GSE's is $206B.  Rational market participants should refuse to invest $500B of capital for $206B of value.  

SUMMARY - insisting on a 10% capital ratio without increasing g-fees (g-fees would need to be approximately 217 bps to attract investment) may be the government's strategy to prevent the GSE's from ever going (pretend) private again.