The components of mortgage rates

In looking at the direction of mortgage rates, it is worthwhile to consider the components. Let’s consider two broad categories: the direct costs of the loan, and the margin on the loan.?

First, the direct cost, which can be considered the ‘cost’ of the loan to the lender.?In the case of a 30 year mortgage (the most common mortgage), the most common ‘hedge’ is the 10 year bond.?The recent rise in 10 year rates can be taken as a given, but why not use the 30 year and be perfectly matched??There are a few reasons for this, but the biggest reason is prepayments with the average tenure in US homes of just about 8 years[1] (until recently – more on this later) matching the duration of a 10 year bond quite nicely.

The next main category is the margin, or what a lender charges in excess of their costs.?While the move in the 10 year was taken as a given, there are a number of drivers to this margin, which we will examine in greater detail, in no particular order as I can’t really work out the orders of magnitude, but I think I have all the ‘big’ ones.?

The margin reflects the price of risk.?Lenders take a myriad of risks in making these loans, for which they should be compensated.?The SFR market has had more than its share of uncertainty recently[2], but there are also all the ‘regular’ risks of underwriting, loss, fraud risk, etc.?There is even risk in hedging the prepayments. As indicated above, the average tenure had been steadily increasing over the last 10 years but dropped from around 8 years in 1Q21 to 5.87 years in 1Q22[3].?

A large buyer is no longer buying (and may turn sellers).?The Fed started purchasing mortgage-backed securities in 2009 and by early 2022 held over 2.5 trillion dollars of MBS, which is more than 30% of all outstanding[4]. Just stopping buying is enough to push that margin higher, but the Chief Executive Officer of the NY Fed (which is the Bank that executes the majority of these trades) said the FOMC “could consider” turning seller[5].

Not all market participants are participating.?Another large participant, historically, has been CSFB, although they have not been active lately, consistent with their considerable issues.??

There haven’t been a lot of alternative products.?With mortgage rates so low, most borrowers chose 30 year loans. With no competition from other options, margins could expand.?Adjustable-rate mortgages, rate buy-downs, etc are all products that should be returning to the market soon and may help decrease the margins on 30 year mortgages.?

Of course, for some the only reason to look at rates is to try to figure out which way they will go.?The 10 year seems to be stabilizing, now under 4%, and the Fed may slow down the rate of increases of the Fed Funds. And the margin should decrease as the initial shock wears off and participants adjust to the new realities (and return to the market in the case of CSFB).?The Fed turning seller of MBS though is a real concern, but I think they are savvy enough and sufficiently aware of their position that they will be cautious and take a considerable amount of time.?So, net net, I’d expect mortgage rates to continue to moderate and even come down a bit, mostly due to the margin.?

[1] https://www.attomdata.com/news/market-trends/home-sales-prices/attom-q2-2022-u-s-home-sales-report/

[2] https://www.bloomberg.com/news/articles/2022-10-10/here-s-how-weird-things-are-getting-in-the-housing-market

[3] ?https://www.attomdata.com/news/market-trends/home-sales-prices/attom-q2-2022-u-s-home-sales-report/

[4] https://www.msci.com/research-and-insights/insights-gallery/mbs-owned-by-the-federal-reserve#mbs-owned-by-the-federal-reserve

[5] https://www.reuters.com/markets/us/feds-williams-mbs-sales-could-be-an-option-down-road-2022-05-16/