Stephen Young
Analyst · Raymond James
Sure. Michael, yes, just maybe kind of give you some explanation of where -- where we think we're headed on margin, and maybe I can answer some of those questions in the middle of that. As you mentioned, we had higher accretion than we expected. And really, a couple of things around that. We saw the highest accretion in July and then August and September, it kind of tailed off a little bit and really due to some early payoffs of 2020 and 2021 vintage loans that had kind of 3 handle coupons with these big discounts that sold. So those are not economic decisions, but there -- I mean, there are economic decisions in the fact that they sold, but typically, you keep those coupons. Also, we had a 29% decline in PCD loans this quarter. And of course, those have larger marks. So anyway, all of that, we look at prepayments are really not outside of our scope of what we thought. It's just that some of the vintages were different than we thought and therefore, had bigger discounts. So having said all that, as we think about the guidance for NIM going forward, really not a lot of change, a little bit of change, but not a lot. We talk about the size, the assumptions of the interest-earning asset size. The second is our interest rate forecast. The third is loan accretion and the fourth is deposit beta in an environment where rates are going down. Interest-earning assets, we've been saying $59 billion for quarter 4 average. That's no change. For full year 2026, we're looking somewhere between $61 billion and $62 billion. So that's kind of a mid-single-digit growth. Rate forecast, last quarter, we had no rate cuts in our model. This quarter, we're thinking we get 3 rate cuts in '25 and quarterly rate cuts, 3 more in 2026, so that we would get 150 basis point cut in total and get the Fed funds at 3% by the end of '26. That seems to be somewhere where the market is. As it relates to the third assumption, loan accretion, based on our models, we expect loan accretion this quarter for the fourth quarter to be somewhere in the $40 million to $50 million as expected prepayments fall. Our October accretion so far is in line with these expectations. And as I mentioned, August and September came down pretty ratably. So I think -- I think that's a good run rate to use. For 2026, we did certainly pull some forward in 2025. So we expect instead of $150 million of accretion, we're looking at about $125 million based on our prepayment forecast. But of course, it can be lumpy based on these vintage loans. The last part is deposit beta. For the first 100 basis points of cuts, our deposit costs came down about 38 basis points from 2.29% to 1.91%, so 38% beta. In our 2019 to 2020 easing cycle, our deposit beta was around 27%. So our expectation is with the growth plans that our deposit beta would look a little similarly to 2019, 2020 to 27%. Maybe we get to 30% over time with a lag, but I don't think it will be as high as 38%. So based on all those assumptions, we'd expect NIM to continue to be in the 3.80% to 3.90% range with the step down in accretion this quarter in fourth quarter and for 2026 for to be in that range, 3.80% to 3.90% as we kind of move forward. But one of the questions you asked was our pricing dynamics. Our new loan production rate for the total company this quarter was 6.56%. If you look at Texas and Colorado, that new loan rate was 6.79%. So it's a little bit higher in Texas and Colorado, but it's in total, it is 6.56%.. So I know you have a few questions, a few puts and takes, but that's some guidance for you.