The real reason mortgage rates are falling at a record pace

(Much of the following is adapted from yesterday’s coverage, but with additional charts and examples. As for the push pins, if you got the message yesterday, today’s version is just a little more or less the same.)

Mortgage rates have recently been all over the map, both in terms of movement and variation between lenders. The disconnections are so large that they require more in-depth explanation.

We quantify “great”. Best-case rates for a conventional 30-year fixed rate are down about half a point from last week on average and more than a full point below their mid-June highs.

This is a exceptionally fast fall! In fact, it is somewhat faster than the decline seen in the yield on 10-year Treasuries, a frequently used benchmark for mortgage rates.

20220729 nl1.png

That makes it even more exceptional because 10 years almost always move faster than falling mortgage rates. Here’s an example of the last time rates fell from a long-term high:

20220729 nl2.png

So why are things different this time around, and what about the extremely sharp decline in the past few days?

A caveat: Answering these questions requires a guided tour of some high-level concepts involving the bond market. If it’s all a bit confusing, that’s normal. We will agree on the more esoteric things in the following six steps. These provide the foundation for understanding what is going on.

Step 1: Mortgage rates are mainly based on mortgage-backed securities or MBS. As lenders originate mortgages, these mortgages can be “turned into” MBS and sold to investors wishing to earn interest on the mortgage debt.

Step 2: MBS has a coupon, which is the official rate paid by a bond. Other bonds, such as the 10-year Treasury bill, also have coupons.

Step 3: MBS have a price. Same story for a 10 year treasury bill. This is the price an investor pays to own $ 100 of that bond. A price can be higher or lower than $ 100. If I pay more than $ 100 for a bond with a hypothetical 4% coupon, I technically make less than 4% yield because I paid more upfront. Coupons are periodically set in stone and the bond market moves by changing the price it pays for that coupon. The combination of price and coupon allows investors to know the actual return associated with a bond. So when you see 10-year returns moving all over the place every day, the coupon has never changed. Just the price.

Step 4: Unlike Treasuries, which change coupons only once per quarter, MBS coupons are offered in half point increments (e.g. 4.0, 4.5, 5.0, etc.) and are always available for purchase / sale by investors.

Step 5: In other words, investors have CHOICES TO MAKE when it comes to which MBS to buy. Investor demand for a given coupon can rise and fall for a variety of reasons. In general, however, when investors think that the rate market in general has peaked or that rates will continue to fall, they prefer to buy the lowest possible MBS coupons. As demand increases, the lower coupons gain value more quickly.

Step 6: Mortgage lenders have choices to make when it comes to choosing an MBS coupon in which to place their newly originated loans. There are limits here. Each MBS coupon is limited to mortgage rates between 0.25% and 1.125% above that coupon. for example, a 4.0 coupon MBS is like a bucket that can only hold mortgages with rates from 4.25% to 5.125%.

Here’s how all of the above unfolded recently:

Falling rates and increasing demand for refinancing can hurt investor returns.

Investors buying MBS see rates likely peaked in June and have since fallen. Gloomy economic data added to this momentum this week. If rates continue to fall, borrowers who recently closed will start refinancing, effectively repaying investors too soon, before investors have had a chance to earn a lot of interest. In other words, early refinancing costs investors.

Investors try to get ahead of the game by buying lower coupon MBSs (which are less likely to be paid off upfront due to falling rates). This increases the value of those coupons.

Why on earth should you worry about a lower MBS coupon rising in value?

It may interest you more than you think. Mortgage lenders base their rates on several factors, but the MBS coupon price associated with those rates is the most important consideration because most lenders convert your loan into MBS to sell to investors.

It sounds simple enough, but again, MBS coupons can contain rates between 0.25% and 1.125% above the coupon (remember 1.125% … we’ll use that in a moment). So a mortgage lender has a choice to make. Any loans they close can be placed on one of the two MBS coupons.

Lenders also love to make money.

We talked about how investors change their buying habits to make more money. This directly leads lenders to change their selling habits to make more money! Let’s use a real-world example from this week to bring us back out of the rabbit hole:

  • Rates were in the upper mid-range of 5% last week.
  • A rate of 5.75% is too high for an MBS coupon of 4.5.
  • A rate of 5.625% is NOT too high for a 4.5 coupon (4.5 + 1.125 = 5.625)
  • Investor demand made MBS prices of 4.5 coupons higher than normal compared to prices of 5.0 coupons.
  • This meant that the lenders selling MBS actually made more money selling a 5.625% loan than a 5.75% loan!
  • Note: The rate still takes into account in the profit equation. In other words, with a sufficiently high rate, a 5.0 coupon can be more profitable than a 4.5 coupon, but not until it moves at least 0.25% more. Here’s an example comparing rates, coupons, and relative prices from a lender who earns the most by giving you 5.625% versus 5.875:

20220729 nl6.png

Going down to the next MBS intersection, we find a similar scenario with 5.125% and 5.375%.

20220729 nl5.png

Perhaps more important than the fact that 5.125 is roughly as profitable as 5.375 to your lender is the fact that there is only 0.79 in the price separating 5.875% and 5.125% (103.13 – 102.34).

A 0.79 improvement in MBS price is very big, but it’s the kind of thing that happens over the course of a single day every now and then, and over 48-hour time frames quite often.

This means that 2 good days in the bond market could bring a 30-year fixed rate quote down a staggering 0.75%, a relatively unheard-of amount for such a short amount of time.

A note on points: The much smaller than normal profitability gap between certain rates can be paid in advance in the form of “points”. Using the example above, a lender earns the same amount of money by giving you a rate of 5.875% with no points or 5.125% with 0.79 points.

Which is better? It depends on the borrower. It would take about 17 months to recoup that extra initial cost through monthly savings, and the rates could easily drop enough between now and then that it makes a lot more sense to refinance. Comparison is not the point (no pun intended).

The point is that “points” are more prevalent and more valuable than normal, a fact that helps us reconcile the seemingly large differences between lenders and rate quotes that at first glance may seem too low to be real. Thanks to the points, there were a lot of 30-year fixed rate quotes at 4.625% by Friday afternoon this week.

Other news this week:

Many of the links below have the full history of many of the week’s biggest developments, but to be sure, Thursday morning’s GDP data was the week’s biggest market driver for rates. We also heard from Powell and the Fed on Wednesday, and although the Fed raised rates by 0.75% again, the markets had been pricing it for some time.

Powell moved the Fed off the path of clearly telegraphed rate hike amounts. Market participants have debated whether this was rate friendly or not, but best of all is likely that the Fed will be guided by upcoming data when it comes time to decide the size of the September rate hike. That valuation is perfectly in line with the relative calm in bond markets for an afternoon of Fed announcements.

Leave a Reply