tag:blogger.com,1999:blog-4815867514277794362.post1639794973646912422..comments2024-03-08T06:18:28.125+11:00Comments on Bronte Capital: Modelling Fannie Mae and Freddie Mac – Part IIIJohn Hemptonhttp://www.blogger.com/profile/03766274392122783128noreply@blogger.comBlogger7125tag:blogger.com,1999:blog-4815867514277794362.post-34416114514659193522009-08-14T19:43:20.285+10:002009-08-14T19:43:20.285+10:00go on Jono go on to your series
distract mildly.....go on Jono go on to your series<br /><br />distract mildly..owlnoreply@blogger.comtag:blogger.com,1999:blog-4815867514277794362.post-51155846243027475272009-08-14T18:44:00.096+10:002009-08-14T18:44:00.096+10:00I thought about it and you are probably right with...I thought about it and you are probably right with regards to the older limits. That said my colleague bought his town house beginning of 2008 (!) for half a million from the developer using a 500 dollar (sic) down payment. Well, nothing like coming out of college and getting your first paycheck! At that time he told me he stole one from the developer.<br /><br />In any case, this anecdote reminds me that you don't show graphs for 2008 and 2009. One would have expected more prudence, but this and the recent buying frenzy in the "low end" (<700k) tells me that the government took over the easy credit business. <br /><br />I apologize for looking for holes in your arguments and giving you a hard time. In general I really like your articles.IFnoreply@blogger.comtag:blogger.com,1999:blog-4815867514277794362.post-18583824592585645902009-08-14T18:40:41.940+10:002009-08-14T18:40:41.940+10:00The FT published a similar Moody's chart for S...The FT published a similar Moody's chart for Spain a few months ago. The similarities are evident. But they also provided a line, described as an 'index', that probably shows (assuming some kind of weighting) how badly hooked the banks and cajas are to post-2005 loans. Also of interest in both the U.S. ad here is the rapidity with which the 2008 vintage got airborne.<br /><br />http://ftalphaville.ft.com/blog/2009/05/18/55898/the-pain-in-spain-will-be-felt-mostly-by-the-banks/?source=rss<br /><br />or (if trolling is acceptable)<br /><br />http://ibexsalad.blogspot.com/2009/05/how-to-rescue-cartel.htmlCharles Butlerhttps://www.blogger.com/profile/00486529931043507880noreply@blogger.comtag:blogger.com,1999:blog-4815867514277794362.post-29600324760262603302009-08-14T18:19:20.423+10:002009-08-14T18:19:20.423+10:00Excellent analysis. The best I have read so far, o...Excellent analysis. The best I have read so far, or shall we say, there are not really any analysis out there, but anyway, this is good.<br /><br />I look forward to reading more, and could you please advice me where I could find more information on your Fund?<br /><br />I think if you could explore further two issues, that are very important.<br /><br />Underwater mortgages are 400 BUSD at FNM and 350 at FRE as of June 30 2009. In the latest report FNM states that of these 150 is currently estimated to be underwater with more than 125%<br /><br />Now they had Dec 31 2008 a serious delinquency rate on this part of 18%, or around 28 BUSD. In June 30 they had 25% delinquencies on about the same amount, that is around 38 BUSD.<br /><br />Now we know NOTHING about HOW much above 125% these mortgagees are. What we know is that Fannie and Freddie, when they bought private label, or other than conforming loans, they bought the highest rated tranches.<br /><br />I look at the report some years back, and the Alt-A portfolio then around 300 BUSD had an average FICO above 720 and a M2M LTV at that time of an average 57%. To me is sounds like this actually is the best tranche.<br /><br />I assume that when they bought these AAA tranche, most mortgages actually were originated with low LTV and high FICO scorses, otherwise they would not have been in the tranche. The info about Alt-A given also supports this.<br /><br />Do you have any other information you could combine with this to try to find out amount between 125-135, 135-145, etc.<br /><br />I think, logically, FNM and FRE would have very few mortgages deep underwater. Since it mathematically impossible if they should have it, starting out with very low LTV 2,5 years ago.<br /><br />I think the answer is very important since it will be from this 150 BUSD (FNM) that most of the credit loan losses will come.<br /><br />The other thing which hurts especially FNM is the NPL's in that the forgone income stream from the mortgages is very costly. With a total of 177 BSUD of NPL's FNM on a rolling y-o-y basis is currently loosing, say, 8-10 BUSD in revenues, but still has the cost to finance these mortgages.<br /><br />FRE 77 BUSD is in this respect better, even though they have about the same underwater delinquent mortgages, FRE overall NPL's are lower, which is good for its revenue.<br /><br />I note that, still, 140 BUSD of these 177 BUSD FNM are above or not more than 120% underwater. I'm a little bit confused about this, since on one hand this means the borrower should have a high incentive to start paying again, or risk loosing the house that are currently not or only somewhat underwater, but on the other hand it also points to the fact the FNM does not foreclose even though they loose a lot of money holding above water mortgages.<br /><br />Could you say anything about this?<br /><br />Thanks in advance, will read your blog daily from now on.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-4815867514277794362.post-54994944089869311022009-08-14T13:39:20.840+10:002009-08-14T13:39:20.840+10:00Oh, the expensive places being underwater doesnt m...Oh, the expensive places being underwater doesnt matter. None of the mrotgages in the trad book are abvoe 330K. If a million dollar house with 500K in mortgages sells for 400K then Fannie or Freddie loss equals zero.<br /><br />JJohn Hemptonhttps://www.blogger.com/profile/03766274392122783128noreply@blogger.comtag:blogger.com,1999:blog-4815867514277794362.post-68340269439500663442009-08-14T13:37:36.338+10:002009-08-14T13:37:36.338+10:00I am aware how non-linear the walk-away option is....I am aware how non-linear the walk-away option is...<br /><br />But then the 2004 pool has only 175 billion in it and less than 2 percent delinquency.<br /><br />Its just not going to hurt us. Simply a non-starter as an issue.<br /><br />By contrast - the 2007 pool - which is as scary as anything - has 307 billion left and 6.8 percent delinquency.<br /><br />That can hurt us. If I had really good loan-by-loan data on the 2007 and 2006 pool I think I could get myself really comfortable.. <br /><br />But yes - I worry terribly about my 2006 and 2007 year estimates - and I cover those things off mainly in Part VII...<br /><br />JJohn Hemptonhttps://www.blogger.com/profile/03766274392122783128noreply@blogger.comtag:blogger.com,1999:blog-4815867514277794362.post-18070119097977550052009-08-14T13:28:06.392+10:002009-08-14T13:28:06.392+10:00As you admit, numerical extrapolation is a fools g...As you admit, numerical extrapolation is a fools game. Even knowing a lot about a function and all its derivatives (Tailor series expansion), the error outside of an interval can be very high.<br /><br />That said, you are mainstream/practical enough to ignore the case where pre 2004 starts up-sloping. Many of these might have been refinanced and what is refinanced can't default.<br /><br />Then again a graph like<br />http://www.calculatedriskblog.com/2009/08/american-corelogic-more-than-152.html<br />shows us that a lot of expensive places are underwater. I keep telling my colleagues that they were given a free option to walk. (Non-recourse loans.) So far they insisted that I am talking crazy. But just today one of them whined about him not wanting to pay that much for his house. I read more of such mind changes across the country.<br /><br />Just saying, you have not modeled the cost of the walk away option yet. It is very non-linear and I doubt you can predict it by simple extrapolation.IFnoreply@blogger.com