Monday 7/28 ... Bad Friday, Followed By A Good Monday

Friday saw a very rough day for MBS (mortgage backed security) pricing.  Recall that as the prices of MBS fall, mortgage rates rise.  We had surmised last week that the market was near or at it's 'bottom' (lowest prices / highest rates) of the year and we could see steady improvement from here on out.  These sentiments are always tempered by what we refer to as "headline risk."  This is simply the risk that either a scheduled headline will deviate greatly from expectations or that an unscheduled headline will have implications that are not friendly towards the MBS market.

This was absolutely the case on Friday as the "Durable Goods" report which measures the amount of demand that various producers have for goods came in at a wildly and unexpectedly high level.  These are the risks we run when we choose to float instead of lock.  Indeed the market pared nearly all of the previous weeks gains in one short day.  All hope looked lost, but we turned back to our technical analysis of the market combined with what the fundamental factors were suggesting and STILL saw potentially positive trends in store for MBS.  Thankfully, this assumption turned out to be correct as we've recaptured almost all of Friday's losses today. 

The Numbers

The very best 30 year fixed rates are pushing back down to the 6.25-6.375 range.

The News

• There were no scheduled economic reports today, but the Senate did pass the housing bailout bill over the weekend.  This would allow the government to invest in Fannie Mae and Freddie Mac as well as numerous other potentially beneficial aspects such as modernizing FHA lending.  The fact that this passed the senate so quickly and that they worked a weekend to do it likely contributed to some easing worries among MBS investors.  The victory margin was also reassuring as it was passed by nearly 7 to 1. 
  
• Now the bill goes to the president where it is expected to be signed without veto.
• Another aspect to consider about Friday was the fact that trading was fairly low volume.  This means that any investment firm that placed a large buy or sell order would have had an inordinate impact on the pricing.  Now, with more volume in the market, buyers of MBS are noticeably outweighing sellers which helps bond prices rise and mortgage rates fall.
  

The Call

    Last Thursday we suggested that despite uncertainty about whether or not early June truly represented a "bottom" for the MBS market, we were leaning towards floating.  As the headline risk crushed us on Friday, this indeed appeared to be a very risky call.  But now that we can see how things have played out today, especially considering there is not much economic data driving the trading, it lets us know that the general sentiment was correct.

Whether or not this is a "false positive" or not remains to be seen as the scheduled data for the rest of the week can have quite an impact.  Friday especially is important as it hold the highly important employment report which includes what many consider the most important event for traders on any given month: "non-farm payrolls."  it is basically the most widely watched and most accurate measure of employment we have.  If employment is stronger than expected, it is almost always bad for mortgage rates.  If the economy lost more jobs than expected, it will likely be good for mortgage rates, especially considering recent weakness, the housing bill passing, and other invisible forces at play near the end of the month.

This doesn't necessarily mean that risks favor floating this week.  That will depend on personal preference.  Floating into Friday is like making a bet on a weak economy helping rates.  Remember that analysts at numerous firms are constantly calculating what they feel the actual numbers will be.  These estimates are compiled into an average which is then accepted as the "consensus."  So not only would numbers need to be weak, but they'd have to be weaker than analyst already probably think they are.  Because of this, for the more risk averse, this may mean that locking some time soon might make sense.  Certainly, you'll want to look for rates by the end of today to be as good or better than Friday morning's rates and better yet, thursday afternoon's.  Floating in the short term (under 1 week) is a bit of a gamble, but as far as the mid-term (1-3 weeks), the data suggests things should be slightly more positive than negative.  This can all change at the drop of a hat if any economic data is stronger than expected.  

In general, remember that you'll regret floating when you should have locked a lot more than locking when you should have floated.  in other words, having money taken out of your pocket generally feels worse than missing out on having extra money added to your pocket (just think of the game shows where someone gets out with 200k, only to find out the next answer would have doubled their winnings.  I'd imagine their pain is much less than the person that risked their 200k and lost it all).  Also, locking, in general, is the safer decision because rates can either go up, go down, or stay about the same.  This makes locking a pay-off 2 out of 3 times.  These are not hard numbers, but you get the idea. 

Whatever the case, if you are going to float (and I'm not trying to dissuade you from that with the previous paragraphs), just make sure you keep your eyes on the market.  in general, things to fear include a rise in treasury yields (as MBS loosely follow treasuries) a rise in stock prices (because as money flows INTO stocks it generally flows OUT OF bonds), or an unexpected headline pertaining to the mortgage industry (don't let housing numbers freak you out though).  Either way, the "Professional" version of this blog which is linked at the top of the page contains intraday updates as relevant changes take place.  The writing style and information presented is not as easy to digest, but if you read the "MBS Primer" also linked at the top of the page, it should give you enough information to make decisions if there is not an update on this site.