On Thursday afternoon with less than an hour left to trade, Trump announced he would be directing his representatives to buy $200bln in mortgage-backed securities (MBS). No matter what one's personal opinions may be regarding bold announcements from the President, this one is definitely serious.

How do we know? 

First off, if you don't want to apply any critical thought to the matter, simply consider that the MBS market reacted immediately and forcefully.  In other words, traders have done the mental heavy lifting for you. They wouldn't have been moving billions of dollars of MBS at a time of day typically reserved for playing Angry Birds on the train back to the burbs.

Those who want to dig a bit deeper can do just a little bit of math. Trump said "$200bln." Incidentally, as of the most recent monthly filings, the GSEs (Fannie and Freddie, who would be the entities buying this MBS) have $202.9bln of room on their balance sheets before hitting their current regulatory cap.  In other words, $200bln is a carefully chosen number and not some random, off-the-cuff remark.

Last but not least, the FHFA director has already confirmed it

Does the president have the authority to do this?

In a word, yes. Under the existing regulatory and conservatorship authority, FHFA can direct the GSEs to buy MBS up to the caps set forth in the PSPAs (the legal documents that govern the conservatorship). In fact, this has already been happening over the past 7 months as GSEs ramped up MBS holdings significantly. 

Do the GSEs have the money for this? Where would it come from?

Recent filings show the GSEs have nearly $200bln in cash equivalents (actual cash + short term liquid securities). The money does not come from Treasury or the Fed, etc.

How big a deal is $200bln?

This depends on a few factors, but the reaction in the MBS market is enough to tell you that it matters. Also consider that we can assess the impact of that $50bln of additional purchases over the past 7 months and see what it's done to spreads between mortgage rates and Treasuries.  

Conventional wisdom will have most people comparing mortgage rates to 10yr Treasuries, but that's not the right way to assess spreads these days.  5yr Treasuries are currently a better benchmark and will provide a more realistic idea of what that $50bln in MBS buying did for spreads. Granted, there are some other factors at work that can account for the tightening of mortgage rates to 5yr yields, but during the time that MBS purchases were ramping up, that spread tightened by about half a percent (50bps).

There are diminishing returns to additional MBS purchases as this spread continues to compress, but it's not overly optimistic to think it could result in another 50bps of tightening in the best case.  This means that today's 6.125% mortgage rates could fall to 5.625% without any movement in the broader bond market.

But the true impact depends on the details of how this is conducted. We know the MBS market is definitely already reacting, but that reaction will continue to be volatile and unpredictable as traders rush to price in the unknown impacts of unknown implementation details. It's been very messy so far, and you can expect much larger-than-normal swings without any new provocation.

Just this morning, MBS rallied by more than half a point only to lose the whole thing an hour later and gain half of that back in about 30 minutes.  These were all real trades backed by real volume.

What's this going to do for mortgage rates in the near term?

Not as much as you might think based on the MBS movement.  Yes, it's true that rates are based on MBS, but there are other considerations. Volatility costs lenders money to insure against.  The higher the volatility, the less of an impact an MBS rally will have on rates. Case in point, even though MBS prices are up more than half a point from yesterday morning, rate sheet prices are only reflecting about half of that improvement so far today.  Moreover, that's just the average lender. Some lenders barely budged.

This could eventually come out in the wash if volatility stabilizes and the details of the bond buying plan become locked in.

What's the bottom line for now?

This is not Fed-style QE. It's not an ongoing commitment to steadily buy MBS until further notice. That said, it's definitely something and it definitely suggests rates will be left in a better position than before. 

Estimates will vary quite a bit on what the total eventual impact might be. Use the example of the past 7 months as your guide (but keep the diminishing returns in mind). A mere 50bln in additional MBS buying led to a 50bp (0.50%) tightening of mortgage rates to Treasuries over the past 7 months. Even after diminishing returns, it's not outrageous to think another 50bp is possible.  That said, it's unlikely that it would be too much more than that due to the finite nature of this bond buying commitment. We also don't yet know what the timeline of purchases will be and whether there will be any unforeseen roadblocks.

Bottom line: it's good, not bad.  Exactly how good remains to be seen.  Don't expect an overnight miracle.  Expect volatility.  Expect certain lender rate sheet changes to NOT fall in line with what you've come to expect over the years.