Joel Kan, VP, Deputy Chief Economist, Mortgage Bankers Association

Navigating the Economy and Mortgage Market in Uncertain Times: How to Prepare for the Year Ahead

The presentation-style session will provide perspective and insight on the current economic environment, how that impacts rates, housing and mortgage markets, and the Mortgage Bankers Association's forecast for the remainder of 2025 and 2026. Additionally, the discussion will dive a step deeper and talk about the impact of technology on mortgage operating costs and productivity.

Introduction by Holly Sraeel, SVP, Strategy and Content, National Mortgage News Live Media

Transcription: 
Transcripts are generated using a combination of speech recognition software and human transcribers, and may contain errors. Please check the corresponding audio for the authoritative record. 

Joel Kan (00:07):
Thanks for having me. Great to be here. I don't know how I'm feeling about going right before the FOMC announcement at two o'clock today Eastern time, so this could be really good or really bad. I'm just kidding. No, but again, thanks for having me. It's always a good opportunity for me to get in front of lenders and vendors and to get feedback from you all. I will be available for Q&A after, so I'm going to go through my presentation and then we'll take questions later. Anyway, the way we usually break this down is I'll start from the highest level view of where we think the economy is and where it's going, getting down into interest rates. Obviously, we'll talk about the Fed a lot along the way, and then get into some housing-specific metrics and data that we look at.

(00:57):
For the sake of this group and for the occasion at this conference, I also want to get into some of our profitability and productivity metrics, and then try to connect that with the technology piece. I know there are lots of more detailed sessions following me, including at least two, and then there's also a regulatory one as well, which I think hopefully this will be a good opening act for. That being said, I'm going to start with the broadest view of where we think the US economy is and where it's going. Really quickly, 2025 is just not as strong of a year as '24. We had a 2.5% growth rate in 2024. What that means is, if you think of the US average growth rate at around 1.8% to 2%, '24 was a pretty good year.

(01:50):
We came into '25 with some momentum and thought that we would get a pretty similar year, but as the data have been released and as we've thought about what's impacting the economy, we've marked down that forecast to about 1% for 2025. You would probably expect a lot of that has been driven by this slowdown in consumer spending, maybe impacted by consumer confidence, and then also, as a result, some weakening in the job market, which we'll get to. One big topic that has been all over the news this year has been tariffs, and that is also impacting our view. The chart on the left is a picture of where tariffs, or the effective tariff rate, was at the end of '24.

(02:44):
And if things go into or stay in place as they were on August 1st, what the impact would be as far as the effective tariff rate goes. What does that mean? Tariffs basically are a tax or a levy on goods brought into the US that are being imported. At the end of '24, we have about a 3% to 4% effective tariff rate on all goods coming in. Again, if everything is implemented and stays the way it was as of August 1st, that effective tariff rate would be about 17%. Thinking about it personally, the hit to what you are getting in terms of income is big. Going from 3-4% to 17% is a significant increase in the tariffs that companies are paying on these goods.

(03:37):
Obviously, this varies by industry and by the type of good. Also, which we'll get to in the second chart, is how much of the cost these companies can absorb versus passing on to consumers. Because their inputs are more expensive, how do you solve for that? You either absorb the additional costs or you pass them on to your customers. The chart on the right, with data from the Federal Reserve Bank of New York, shows what companies think they're going to do. About half of them think they'll pass on about 100% of the increased tariff costs to their customers. On the left side, the other extreme is they're going to absorb—I think it's 20-something percent who are going to absorb some of the tariff costs.

(04:29):
What does that do? Obviously, in the middle, you have the range of possible outcomes that companies may or may not want to do. At the end of the day, you either are passing on the cost to the consumer, which makes things more expensive—and they're going to pull back a little bit in terms of spending—or the impact would be on the company. You're either passing on partially or absorbing partially, so that's impacting your bottom line. Companies just get a little bit less profitable, to oversimplify this. One more thing here that I note: I kept saying "as of August 1st" because we've seen this starting and stopping as far as the administration goes—are they going to do it, are they not? Is it a 20% tariff or 10%? And then these negotiation periods.

(05:13):
If you're running a business, it's pretty tricky to try to price things six to 12 months down the line when you don't know how those input costs are going to change. Earlier in the year, you might recall there was a lot of news around people and companies front-loading their purchases of big-ticket items. A lot of that happened in Q1 because everyone wanted to get ahead of the price increases. You have this hit to profitability and income that we think is impacting economic growth in general, but then also these uncertainties in the near to medium term as to how to operate your business or make decisions as an individual consumer.

(06:09):
All that is playing out across the economy in terms of disrupting how you might be budgeting and spending. A lot of that certainly is weighing on the overall outlook. If companies aren't doing as well, how does that show up in the job market? This chart is a high-level summary of the Bureau of Labor Statistics monthly report that most of the analyst community and even the Federal Reserve relies on for the health of the job market. The bars show how many jobs we are adding per month to the economy. The lines are the unemployment rates.

(06:58):
The bars show that as of August, we added about 22,000 jobs. For year-to-date 2025, that average is about 75,000. That's down from about 170,000 in 2024 for those keeping track at home. There was a recent release showing that number is going to be revised down as well to about 80,000 for 2024. All that to say is the level of job growth we're seeing has already been weakening. The job market has shown signs of weakness, and I think we're at a point where it is pretty weak relative to where we were in '24 and especially '23. The headline unemployment rate is that line on the bottom, about 4.3%.

(07:48):
In August, we were in the 3% range not too long ago. The Fed's dual mandate is for full employment and price stability. How do they figure out what full employment is? It's assumed that it's around 4%. That's kind of where they want to be. We're a little bit past that, so it's a little weaker than they want it to be. The other line I'm showing is something we're following because it's not the headline unemployment rate, but it adds workers who are discouraged or working suboptimally because they're working part-time for economic reasons. They would rather be working more, but they just need to supplement so they're working part-time because they can't find the right full-time job.

(08:40):
Those indicators all add up to illustrate a job market that is slowing because of the slowing across the economy. Quick side note on this: I know there's been some controversy around the reliability of these data, and if you look across other data sources, the Atlanta Fed has a really good wage growth tracker. The wage growth tracker is showing that wage growth is cooling pretty quickly, both for people staying in current jobs and for people switching. To me, that indicates weakening in the job market because when the job market is strong, you get more benefit from switching because there are opportunities out there. Clearly, that's not happening.

(09:29):
I mean, they're not low by any means; we're still above where we were in the previous decade. But the pattern that wage growth is declining and these metrics showing weakening in the job market means that the Fed is seriously taking this into account and figuring out the next move. Markets are expecting at least a 25 basis point rate cut today because of that weakness. This is another data source from the New York Fed showing AI-related positions and hiring decisions.

(10:16):
The short version is it shows you what positions are most popular for AI implementation, and on the right is the more important picture: how that's unfolding. About 45% or so for the service industry say they're retraining their current workforce. If you look down the left side of that chart, you can see it's either less hiring because of this implementation or layoffs in some cases. This is also something to consider going forward. Will this show up more broadly across the job market? This is a wait-and-see, but clearly, it is starting to impact some hiring decisions. There has also been discussion around the unemployment rate for younger workers being higher because entry-level positions are being filled by technology or someone becoming more efficient because of the technology they have.

(11:34):
As far as the consumer goes, the job market is weakening, so how's the household doing? Consumer household finances are doing okay for now. Debt service ratios and household financial obligation ratios are still at a reasonable level, but some households are clearly being impacted by student loan debt repayment that started this year. This chart shows we're at a $1.6 trillion bill for the country and it shows you the age range. It's not just younger borrowers; it is across all age groups. The repercussions here are an income shock because people are having to repay these loans after a few years of forgiveness.

(12:28):
But then also there's a credit access impact because those delinquency rates have come up, and you're seeing a hit to credit scores. When you think about housing, people have to qualify based on credit score. It is a big piece that is showing up more and more when we talk to originators. Inflation was the trendy headline for economists; this was the big piece because we had a period of really high inflation. The Fed had to cool the economy down quickly in 2022. The Fed funds rate was close to zero; they raised it to the 5% range quickly to try to bring some of this back in.

(13:21):
They've gotten us some of the way there. This is the second half of that dual mandate, price stability. They estimate that we should be around a 2% target for things to be stable. Inflation matters because if you are paying higher prices on things you have to buy, you are going to change your spending decisions. If it's 2%, we sort of assume that's normal. When it gets to 9% like we saw back in '22, then you really have to sit down and budget. We're not quite at that 2% mark; we're still around 3% or so. Some of that is being held up by the tariff impact.

(14:11):
The second thing I'm not showing here is expectations. Consumer expectations of inflation are actually still pretty elevated. They're expecting inflation to stay a little bit higher, which impacts future decision-making. Thinking about those two—the unemployment rate at 4.3% while the Fed wants it close to 4%, and inflation not quite at 2%—where do we end up? They either have to let the economy run hot—if they cut rates, they're letting inflation run a little hotter than they want—or if they want to get ahead of inflation and keep rates restrictive, the unemployment rate will come up. Our forecast is for the job market to continue to weaken. We expect about a 4.8% unemployment rate by the first half of 2026.

(15:03):
Recent communications from the Fed have shown that a rate cut is on the table, and market odds have a 90-plus percent expectation of at least a 25 basis point rate cut today. We do agree that they'll cut because the level of the Fed funds rate matters. We're sitting around 5.25% today. Even if they cut three or four times, we're still in the 4% range, which is still restrictive. They have some space to cut rates to help the employment picture while staying at a level where inflation is hopefully being dealt with, assuming we don't get an upside surprise from tariffs. Baseline expectation from us is a 25 basis point rate cut today, another 25 bips maybe in December, and two more rate cuts in the first half of next year. What does that do for mortgage rates?

(16:47):
The last three or four weeks we've seen mortgage rates come down by about 30 basis points in anticipation of this. We think the 10-year Treasury is not going to go that much lower. We're sitting at around 4.05% last I checked. You have to check these things right before you get on stage considering how quickly it changes. Our forecast is for the 10-year to stay in the 4.2% range. Things move quickly, and we've all seen 10-20 basis point changes in a week. Yes, it is a pretty flat rate forecast, but again we could see a 4% or 4.4%. It depends on the data and the communication from the Fed, market players, and the administration.

(17:35):
Expect some rate volatility around the 10-year, but kind of close to that 4% level. The short end of the curve is coming down because the Fed is lowering rates over the next 12 months, but the long end is held up by overall uncertainty around US debt and the deficit. We have a large budget deficit that needs to be funded. And now there is another possibility of a government shutdown. Remember last year we also had our debt downgraded by Moody's. All these things are playing into the longer end, and that's going to keep rates a little bit higher—at least the 10-year higher for the long run.

(18:23):
For mortgage rates, we have our forecast going down to about 6.2% by the end of next year, or mid-to-low sixes in 2026, maybe a low end of 6.2% in '27. Our survey today had the 30-year fixed averaging around 6.4%. Like I said, things could go up or down quickly. We're thinking the low sixes to 6.5% is that range we're going to be in over the next 12 months. If we get a little bit more clarity around the current Fed easing cycle, then that might bring volatility down a little bit.

(19:11):
That might bring the 30-year fixed-to-10-year Treasury spread in a little bit, because some of what's keeping that spread wider is that volatility component and prepayment risk. Because rates can also come down, MBS investors want to be compensated for that prepayment risk. We think that's keeping that spread a little bit wider at 220-230 basis points. Typically we've run at 180 or so historically. That volatility has given us opportunities, especially on the refi side. Our survey today had the 30-year fixed at around 6.39%. The point here is that even though average rates are not in the fives yet, these pockets of rate drops have helped on the refi side.

(20:20):
Recent originations have been in the 7-8% range, and those borrowers stand to gain even with rates in the mid-to-low six range. That chart shows the annual changes in our refi index whenever rates dipped over the last year. Maybe just as significantly, on the right, you can see the average loan size on a refi moves pretty drastically because if you have a large loan size, even a 20-30 basis point rate incentive might be beneficial. This was done before today's release, but our refi average loan size went up to about $460,000, which is the highest ever in our survey. Again, borrowers with larger loan sizes are just more sensitive to rate movements.

(21:09):
This is a new development: the ARM share (adjustable-rate mortgages). The ARM share for purchase is on the left; the ARM share for refi is on the right. The purchase chart is not that surprising; in a purchase market, you have a higher ARM share because that's where borrowers typically benefit the most. Most of these are 7/1s or 7/6s. On the right, the ARM share of refis is coming up pretty quickly. In this rate environment, if we think rates are going to be in the 6.2% range, it actually does make sense for someone to refi into a 7/1 because seven years is a pretty good horizon to get that lower rate for. You can either refi along the way or let things float.

(21:54):
Purchases are still running 15-20% above where we were last year. That's pretty good news. More inventory across the country is helping purchase demand even though rates are still in the sixes; people are having to move for life reasons and it's better than a 7% rate. When you're buying a house, rate isn't the only decision; there are many other factors. Finding the house you want is a huge one. Inventory is loosening in some parts of the country and in other parts not so much. This is impacting housing prices, which then also helps with affordability—or not—depending on where you are.

(22:50):
How do you read this? Bottom to top is lower home price growth to higher home price growth. Left to right is less inventory to more inventory. Top left are geographies with low inventory and high home price appreciation—still really competitive buying markets. Bottom right is growing inventory but also downward pressure on home prices. I can't think of scenarios for the bottom left—maybe a town gets wiped out and not rebuilt—and the top right might be more speculative: high inventory, high prices. Our forecast goes for about flat home prices because you saw half with increases and half with declines.

(23:45):
The home price outlook is more important at the state and local level now than nationally. Even though we're looking at flat prices, the growth leading up to the last few quarters still shows up in high levels of home equity. Home equity lending is certainly an opportunity for lots of lenders. As far as originations go, we do expect a gradual increase in purchase originations given the freeing up in inventory, the rate environment, and that demand from the last few years. Obviously, there's been some uncertainty around the economy weakening, so expect that to offset some of that.

(24:34):
Refis are an opportunistic environment. Put those together and you have about an 11% increase in originations in '26 for units. It's 12% growth in units but still pretty low relative to where we've been historically. A lot of that front-loading happened in 2020 and 2021. Lots of loans still have sub-6% mortgage rates. I think 80% of all mortgages have a 5% rate or lower. All right, I kind of used up too much time on the first few, but I'm going to try to get through these hopefully coherently. For profitability, what you're seeing is production income per loan for a sample of about 300-400 IMBs.

(25:27):
Q2 had a 25 basis point profit per loan, about $950, after two or three years of really challenging times. 25 bips in Q2 was driven by volume coming up but also expenses coming down. We talked a lot about efficiencies using technology—yes, you have to spend to play—but at the same time, the industry has been able to streamline around other cost-cutting measures. Unfortunately, one was hiring, but then also just around the edges—where can you cut costs? How do you streamline the operation? The Q2 number was a pretty sizable drop where the cost to originate was down by $1,600 per loan over the space of a quarter. Part of that was volume, but either way you look at it, that's pretty good news.

(26:26):
This was just a decent pickup in volume, but then a pretty good savings cost-wise. Employment has stabilized a little bit. Industry employment has kind of flattened, so companies are now trying to leverage what they have both from a personnel standpoint and from a technology standpoint if we do get a bigger pop in volume. The productivity numbers are important; even in '24, a challenging volume time, productivity was picking up. Again, part of this might be because of overall streamlining, but clearly, with the amount of technology investment, some of that is starting to show up in more efficiency.

(27:26):
Mike and Mike also talked about pull-through and quality. There's a big cost when the loan doesn't close. Unfortunately, we are seeing slightly lower pull-through at least through 2024. We've been hearing that part of this is a qualification challenge—it's just more challenging for lots of borrowers. Part of this is also the overall homeownership cost environment. Taxes and insurance are a massive hindrance for lots of homeowners. We're hearing that sometimes you get to the closing table and someone looks at the T&I estimate and they decide it's not for them. That's a consideration when it comes to pull-through.

(28:18):
Takeaway: pull-through is coming down, and that is something people are watching going forward when thinking about how to improve the quality of the app. They talked about VOA/VOI components so that your pull-through isn't falling. Technology expenses: '23 was the recent peak for that, and in '24 things came down a little bit. All that to say is it's been higher. People are talking more about it. I would wait to see what the first half of '25 brings, but generally speaking, there is a higher focus on technology. Corporate admin covers vendor management, risk management, all that stuff.

(29:05):
It implies folks are utilizing more technology; that's why they need the compliance and risk management components. Last week I was at an HR event and learned that even in HR functions, they're using a lot more technology and AI for resume screening, internal tools for the employee handbook, how to get time off, et cetera. More and more is showing up here in that corporate admin piece. This chart is a little bit hard to see. Even though the tech piece is a small sliver here, this is what I really want to talk about. This is important because we often look at the tech line and sort of judge based on that.

(30:01):
But as the previous panel talked about, that has a spillover effect because the tech piece impacts the rest of the process quality. Mortgage is a production line, so the ROI may not show up here specifically in dollars, but it does show up in terms of a higher quality process. It does help the sales team. It is the impact across the life stage of the loan and not just focusing on that one component. The reason it's so small here is because of the way it's defined—this form was not meant to be tech-focused; it's a financial reporting form. The tech costs are spread out over all those other categories.

(30:56):
Really quickly, this is from Fannie Mae—a really good survey. Lots of priorities are for business process streamlining. The AI component—this group is well aware of where we're going with this. I would say the one thing we are tracking is the compliance and the vendor management piece of this because we have 50 states now wanting to pass some kind of AI legislation. I know the California bill did not go through last Friday, but if you're operating across the country, whether you're a lender or a vendor, there are lots of these to be aware of. I know you guys are going to talk about that later today as well. Sorry, I had to get through the last ones quickly, but I'm happy to take questions later during the networking break. My email is there; I do respond to everyone. Thanks a lot and enjoy the rest of the conference. Appreciate it.