If I go here: http://www.freddiemac.com/mbs/docs/delinquencyrates_021010.pdf I can get the Freddie Mac delinquency for fully amortizing ARM's by vintage year and coupon rate.
The years 2005, 2006, 2007 should be the predominate years applicable to MFA. There is a large variance but if I drill down to the 5 & 5 1/2% coupons it looks as though 7.5% could be a reasonable percentage of expected prepayments. The 2009 10K lists principal amount MBS at $7,292M and carrying value at $7,665M or a difference of $373M. 7.5% of that is $28M or a book value hit on 280 million common shares of about $.10 per share. Taking the 101.3% cost basis of the Agency MBS per the press release and the resulting 1.3% loss from prepayment times 7.5% of $7,292M results in a $7M premium hit or about $.025 per share. The other hit is the reinvestment loss on the $547M early prepayment. Taking 4.0% of $547M which may be low from the marginal unhedged portfolio because the cost of hedged money is already commited is $5.5M per quarter or about $.02 per share per quarter.
My questions to the board: Does Freddie & Fannie have the right but not the obligation to purchase these securities after 120 days of deliquency. If that is true what has been the historic average when these loans have been in fact repurchased. Is it 180, 270 360 days? This would give us a clue as to how much acceleration these repurchases cause and how much would normally be reinvested(and hedged) without this event.
The other thing to consider is that our risk goes down from this delevering at about the same time the FED withdraws from its MBS purchase program.
Comment, corrections welcome.
It's amazing how ridiculous people are on these message boards. It's also amazing how someone can protray themselves as knowledgeable when the rest of the crowd is ignorant. I suggest you people listen to the company's earnings calls and read the SEC filings. RMBS are not new products. Defaults, premiums, discounts, taxable income, prepayments are not new terms. If you do not understand the risks of owning a MBS REIT, own a CD.
I totally agree with what you said & notice the price of the stock today looking in the review. Last year was a strong year for MFA their net earnings is so much better than last year; Unlike CMO and a whole cast of newer RIETS showing up on the seen, a stand out stock, LTC bought back 10% of its own stock to turn there net for the year the same as the year before per share. MFA management buy back there own stock from time to time. too! I can not remember a time MFA was below $5 a share for very long, even when they were not showing a net profit at all, more amazing was they paid a dividend through the whole bad times deal.
MFA has dropped below $5 a share only briefly, when they were not earning any profit. Good insight on the details of the play the play. MFA management still buy there own stock & own it right?
"My questions to the board: Does Freddie & Fannie have the right but not the obligation to purchase these securities after 120 days of delinquency."
Let me shed a little light on this. Seven years ago the GSE's routinely bought out NPA 120+. Then they got really silly with underwriting and leverage and capital levels and stopped buying them out until after an actual final foreclosure was adjudicated because it helped them to appear solvent to just keep the NPA 120+ loans off their books. In January an accounting rule change removed the "better capitalized appearance" advantage that keeping an NPA 120+ loans off their books had been affording them. Without that advantage it now behooves them to go ahead and buy out each and very NPA 120+ rather than masochistically continuing to pay interest on them. They have always had the "...right but not the obligation to purchase these securities after 120 days of delinquency." What changed with the January accounting rule is that now they have a motive for buying them out, and they have announced that they intend to do just that, and to keep buying them out promptly as they turn 120+ from here on out.
Thanks for the excellent reply! Its not surprising I guess that they would do something so uneconomic for appearance sake. So now in the future MBS investors will be more concerned about the quality of their holdings because prepayment speeds will become more sensitive to non performing loans.
I've got another question for you and forgive me for trying to take advantage of your insight/knowledge. MFA indicates that their non agency holdings have "structural credit enhancements". What do they mean by structural credit enhancements? Is that the PMI payoff for the remaining unpaid balance over 80% of original appraised value or do they mean there are tranches in the MBS pool that are subordinate to MFA's position? Thanks again.
Your reply got me to thinking that the market value of GSE mortgages could be futher impaired beyond the effects of this one time prepayment notice.
If the GSE's continue to flush out the past due loans after 120 days instead of 24 months investors won't be able to ride those 20 months of government support and higher CPR's would have to be modeled for non performing loans. It would make investors pay more attention to substandard pools due to much higher prepayment speeds.
Thanks...I didn't realize it was that long. So on average we probably have an extra 10 months(24-4)/2 of hedging expenses that could potentially be lost if the funds can't be properly & promptly reinvested.
so from your numbers... if the book value at the end of january was $7.61 then
the revised book value would be $7.24
would you say this calculation is correct according to your numbers?
No one is saying they will earn less money this year or next. Those dividends look solid to me & boy they came back from not making any net profit & were selling above $5 a share, back then, most of the time.
"would you say this calculation is correct according to your numbers"
No you can't add up these numbers to come up with this. Book value is a function of:
(1) the market value of the mortgages
(2) interest income (-) expenses and dividend payments.
(3) gains or losses from the sale of mortgages
After the GSE announcement last week the market value of all GSE mortgages dropped on the open market due to the expected early prepayment to par or 100 for the percentage of those loans(my estimate of 7.5%) that were 120 days past due. So if those mortgages had a value of 105 they dropped to say 104.5(uncalculated number) to account for ON AVERAGE the early repayment of those to be prepaid at par. MFA's book value dropped right then and there because now their mortgages on average are not worth as much in the open market. MFA lost market value equal to the difference between the 105 and the 100 they will receive on those mortgages(estimated at 7.5%) that will be repurchased early...this is the $.10 per share number I referred to. MFA still paid 101.3 for these mortgages regardless of this prepayment announcement.
When MFA gets paid at 100 for the loans that are past due they will have to writeoff 1.3% of those prepaid loans which is the $.025(not $.25) per share that I referred to. This is a taxable event and their dividend will be adjusted accordingly either in the second quarter or spread throughout the rest of the year so as to not payout more than REIT taxable income for the year 2010. Now we may not see the effect of the entire $.025 in 2010 because they projected some prepay at some point in time either in the remainder of 2010 or later years. So in the end its a timing issue when do I get prepaid for my past due loans...now or in the next few quarters. If its later than now I get to earn my spread after hedging expenses during those quarters. If its now then I loose my spread and I still have my hedging expenses. This is my estimate in LOST INCOME of $.02 per quarter if the funds can't be immediately reinvested.
So in this example MFA lost $.10 per share in book value...$.075 was a loss of unrealized capital gain that went away(105 down to 101.3) and $.025 was a realized taxable loss (101.3 down to 100).