The REconomy Podcast™: Where are Mortgage Rates Headed and What Makes them so Hard to Forecast?

In this episode of the REconomy Podcast™ from First American, Chief Economist Mark Fleming and Deputy Chief Economist Odeta Kushi discuss why mortgage rates are notoriously difficult to forecast and how post-pandemic dynamics are influencing the direction of mortgage rates in 2023.

Don’t miss a single REconomy episode, subscribe today.

Listen to the REconomy Podcast™ Episode 64:

“Since 1972, the 30-year, fixed-rate mortgage has, on average, remained 170 basis points higher than the 10-year treasury bond yield. That spread in the month of April was about 290 basis points. In other words, mortgage rates are very high, abnormally high, relative to treasuries.” – Odeta Kushi, chief economist at First American


Odeta Kushi - Hello, and welcome to episode 64 of the REconomy podcast, where we discuss economic issues that impact real estate, housing and affordability. I am Odeta Kushi, deputy chief economist at First American and here with me is Mark Fleming, chief economist at First American. Hey Mark, what question would you say that you're asked the most these days when it comes to the housing market?

Mark Fleming - Hi, Odeta. That's so easy. "Where are mortgage rates headed?"

Odeta Kushi - You know, I kind of figured. I get the same question. And does your answer usually start with, it depends?

Mark Fleming - You know, I wouldn't be an economist if I didn't start with an "it depends" as a qualifying statement.

Odeta Kushi - Yeah, good point. We're not very decisive in our answers usually, especially when it comes to forecasting something like mortgage rates.

Mark Fleming - Well, it does seem like we've been forecasting rates will go up for the last 10 years, right? Stick with a forecast long enough and you might eventually just get it right. Mortgage rates are notoriously difficult to forecast. And, if you had asked me in 2019, when the average 30-year, fixed-rate mortgage was below 4%, if rates could go any lower, I would have probably said no way. And that's without the "it depends."

Odeta Kushi - I do remember, I think I actually asked you that. And you said no way. It's tough. And you know, you would have been very, very wrong because the average mortgage rate declined below 3%.

Mark Fleming - In my defense, who would have thought it possible. A pandemic? In the forecasting business? We call that an exogenous shock, by the way.

Odeta Kushi - Yeah, that's a little bit tough to predict. And, if you haven't guessed it by now, today's episode will be all about mortgage rates. Specifically, I want to discuss the spread between the 30-year, fixed-rate mortgage and the 10-year treasury bond. So first, what does one have to do with the other?

Mark Fleming - Yes, the dreaded spread. First, a quick review. The popular 30-year, fixed-rate mortgage is loosely benchmarked to the 10-year treasury bond. The 10-year treasury bond is often known as the risk-free benchmark for financial transactions worldwide. U.S. bonds backed by the full faith and credit of the U.S. government are widely considered the safest investment in the world. Assuming we pay those bills on time. Something that's in the news at the moment. Not going there on this podcast episode.  So global demand for U.S. treasury bonds causes their price to go up and down and their yield to change along with it. When the 10-year treasury yield falls - in other words, there's lots of demand for government bonds - mortgage rates follow suit and vice versa.

Odeta Kushi - And mortgage rates are tough to forecast because lots of things impact the global demand for U.S. treasury bonds - economic events, geopolitical events, inflation and more.

Mark Fleming - That's right, for example, the higher the current rate of inflation and the higher the expected future rate of inflation, the higher the yield that an investor would require because investors demand the higher return to compensate for that future cost of money and inflation.

Odeta Kushi - Exactly. That's why we're always saying that the outlook for mortgage rates in today's world is heavily dependent on the outlook for inflation. Now, that relationship between the 10-year treasury bond and the 30-year, fixed-rate mortgage can be quantified. Since 1972, the 30-year, fixed-rate mortgage has, on average, remained 170 basis points higher than the 10-year treasury bond yield. That spread in the month of April was about 290 basis points. In other words, mortgage rates are very high, abnormally high, relative to treasuries.

Mark Fleming - Yes, that's definitely a wide gap. Is there any historical precedent for what's happening with that spread right now?

Odeta Kushi - Well, the last time that the spread was this high was in the 1980s, your favorite decade, but there have certainly been periods of time where the spread between the 30-year, fixed-rate mortgage has widened beyond the historic average levels. Since 1971, there have been 126 months, that's about 20% of the total, where the average spread was at least 200 basis points. This spread usually widens when there's a lot of economic or geopolitical uncertainty, as is the case in today's market, lots of uncertainty around inflation, Fed policy, banking issues and even recession risk.

Mark Fleming - I feel like it's back to the 80s future. Didn't we do a podcast episode where we referenced that recently?

Odeta Kushi - We sure did.

Mark Fleming - It seems like we might want to explain the economic rationale for why mortgage rates tend to be higher than the treasury yield in the first place. In other words, why the spread at all? And, this time, maybe you would like to do the Econ-splaining honors? Too soon?

Odeta Kushi - Maybe a little bit. But yes, I am happy to. We sort of hinted at the reason at the top of the episode when we said that the 10-year treasury bond is known as the risk-free rate. The spread between the 30-year, fixed-rate mortgage largely reflects the risks associated with investing in mortgage-backed securities, otherwise known as MBS. Let's go through those risks, because I think it will really help in understanding why that gap is so wide today.

Mark Fleming - Okay, one more definition before we get into those. Maybe a quick refresher on what an MBS is. Odeta, when you bought your house, where did you get your mortgage?

Odeta Kushi - You know, I love a softball question. I went to a mortgage originator, of course.  

Mark Fleming - Of course, pretty standard, that originator could be a broker or a bank, and they helped you to buy the home and gave you the mortgage. But your originator didn't keep the loan, they probably sold it with other loans that they had also originated around the same time to a mortgage aggregator, like a GSE - Fannie Mae, Freddie Mac, FHA - who bundle your loan, along with others from many different originators into a mortgage-backed security or this MBS that they then sell to investors who might be choosing between buying a mortgage-backed security or that U.S. treasury bond.

Odeta Kushi - Right. So an MBS is an investment product like bonds, while a treasury bond is backed by the full faith and credit of the U.S. government and MBS is backed by a collection of mortgages and investors looking for a steady return might compare similar products, such as government bonds and MBS.

Mark Fleming - They're like substitute goods except the MBS is a little riskier than that government bond.

Odeta Kushi - So we can group the risks into two major categories - prepayment risk and its impact on duration, and credit risk.

Mark Fleming - I think, realistically, prepayment risk is the easiest to cover first because its the most well known. So let's talk about that. One difference between U.S. treasury bonds and mortgage-backed securities is that the U.S. Treasury doesn't pay any principal until the treasury bond matures. You buy a 10-year treasury today, you're not going to get that principal back -until 10 years from now. The investor does get paid interest over time for the term of the bond, and then that principal value at the end, however, mortgage-backed securities work a little bit differently. Homeowners have the option to refinance their mortgage or sell their homes at any time and pay off the loan. That principal is "passed through" to the investor and the investor stops receiving the interest payments associated with that paid-off loan. It's as if the MBS is shrinking over time. But the problem is that when you buy the MBS, you're not exactly sure how quickly that will happen. That's prepayment risk.

Odeta Kushi - So investors need to be compensated for that risk relative to the risk-free alternative. Now, the duration of the MBS is also tied to this prepayment risk. The faster the prepayments, the more quickly that principal is paid back and the shorter the duration of the MBS. The slower the prepayment, the longer the duration. In other words, the sensitivity of the MBS to a change in the level of interest rates is what we're talking about here. The duration of an MBS tends to increase in a rising interest rate environment as there is less incentive for borrowers to move or refinance. We're seeing that in today's market, the prepayment rate of the underlying mortgages typically slows down, which extends the life of the MBS and increases its duration. As a result, the spread between the mortgage rate and the 10-year treasury yield may increase because the MBS investors require more compensation for the increased duration.

Mark Fleming - Ah, yes, the future gets more uncertain the further into it you look. If you think it will be a longer time until you get your principal back, then you need to be paid for that uncertainty risk. Also think about duration risk from the perspective of a bank investing in mortgage-backed securities. In today's market, the duration of a deposit in a bank may be pretty short. We've seen them get very short in some cases, especially given all of that uncertainty. But, as we discussed, in a rising interest rate environment, the prepayment speeds go down and extend the maturity of an MBS. Banks purchase those mortgage-backed securities, they have a higher yield than treasuries, but then they're left with all these long-duration mortgage-backed securities on their balance sheets. And they're faced with what's called a duration mismatch between the short-term deposits and now the longer duration mortgage-backed securities. So maybe a bank investor requires the higher yield to compensate for that uncertainty of duration risk.

Odeta Kushi - That's a really good point, especially in the wake of the recent bank turmoil, a duration mismatch between a bank's assets and liabilities, such as having long duration MBS on their balance sheet and short-term deposits, as liabilities can lead to losses and liquidity risk, but I won't get into that. So we're gonna move right along to credit risk.

Mark Fleming - Credit risk - this is actually the easier one of the two here - is associated with the risk of default. At any point, borrowers can fail to make payments on their mortgage underlying an MBS. From the perspective of an investor in a mortgage-backed security, the implications of a default depend on whether that mortgage-backed security is what's referred to as agency or non-agency. For agency MBS - those are mortgage-backed securities issued buy the GSEs, Fannie Mae and Freddie Mac and Ginnie Mae - they promise the full and timely prepayment of principal and interest. So there's really no credit risk to an investor who buys an agency MBS. But, if not, then the investor also needs to be compensated for the possibility that they won't get all of their principal back. That's a bit different than a treasury.

Odeta Kushi - That's a good point. So, in other words, when a borrower defaults, the agency issuer pays the principal to the investor anyway. That basically ends up being a prepayment event for the MBS investor. However, for those non-agency MBS, investors bear the credit losses from a default. Alright, so these are the risks associated with an MBS that explain why an investor requires a higher interest rate on an MBS than for a treasury bond. But, Mark, can you talk a little bit about how supply and demand dynamics for MBS might play into all of this?

Mark Fleming - Yes, one last nuance. A lot has happened over the last few years. And, in particular, the Fed, over the pandemic, was a large, ready buyer in the secondary market. They stepped in and started buying mortgage-backed securities. They generated lots of demand, increasing the mortgage-backed security price and lowering the yield for investors. This results in lower mortgage rates, right? Because the mortgage rate is tracking with the treasuries. Good, old-fashioned MBS supply and demand 101. Strong demand for MBS bids up the price, reduces the return that the MBS investor expects. So, the spread between the mortgage rate and the 10-year treasury yield decreases, because investors are willing to accept that lower yield to gain the exposure to the mortgage market. They're seeking that safe asset. Now that the Fed is backing out, though, as a buyer, that has also contributed to an increase in the spread.

Odeta Kushi - So slower prepayment rates caused by higher mortgage rates have increased MBS duration. The increase in the duration gap caused by shortened duration of deposits, the possibility of credit default and non-agency MBS, and the Fed backing out of the MBS market have all contributed to the increased mortgage rate spread over treasury rates. So when does this spread return to normal?

Mark Fleming - So much uncertainty. It's possible that we'll see a narrowing spread when the Fed finishes its tightening. Possibly.

Odeta Kushi - That would give investors some more certainty. And, if the spread narrows, there may be some downward pressure on mortgage rates. Because, you know, if we didn't have this large of a spread, the average mortgage rate in April would not have been 6.3%. It would have been more like 5.2%. That sounds pretty good

Mark Fleming - Yeah, but, true, that's one source of certainty, and that's helpful, but the Fed is unlikely to become an MBS buyer again, and arguably the slower prepayment rates and longer duration are more permanent and here to stay. What if the spread doesn't return to the historical average because buyers don't refinance or sell as often as they did in the past?

Odeta Kushi - Well, your take is a lot more pessimistic. Well, if the historical spread of 170 basis points is no more, what is the new normal? Are you going to tell us your mortgage rate projections for the end of the year?

Mark Fleming - Umm, it depends. But wait, wait. Isn't a time to wrap up the show?

Odeta Kushi - Yeah, saved by the imaginary and self-imposed bell. I think it's reasonable to assume that rates will come down a bit in the second half of the year and stabilize if the Fed takes its foot off the monetary-tightening pedal. But, if you've learned anything from today's episode, it's that mortgage rates are notoriously difficult to forecast because they are dependent on so many different factors, so we're just going to leave you with that.

Thank you for joining us on this episode of the REconomy podcast. If you have an economics related question you'd like us to feature in the future, you can email us at We'd love to hear from our listeners. And as always, if you can't wait for the next episode, you can follow us on Twitter. It's @OdetaKushi for me and @MarkFlemingEcon for Mark until next time.



This transcript has been edited for clarity.

Subscribe for Updates

Subscribe to First American's REconomy Podcast Blog for the latest housing market research and analysis driven by Chief Economist Mark Fleming.