Greetings, and welcome to Cullen/Frost Bankers, Inc. third quarter 2021 earnings conference call. At this time, all participants are in a listen-only mode. A question and answer session will follow the formal presentation. If anyone should require operator assistance during the conference, please press star zero on your telephone keypad. As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, A.B. Mendez, Senior Vice President and Director of Investor Relations. Thank you. You may begin.
Thanks, Rob. Our conference call today will be led by Phil Green, Chairman and CEO, and Jerry Salinas, Group Executive Vice President and CFO. Before I turn the call over to Phil and Jerry, I need to take a moment to address the safe harbor provisions. Some of the remarks made today will constitute forward-looking statements as defined in the Private Securities Litigation Reform Act of 1995 as amended. We intend such statements to be covered by the safe harbor provisions for forward-looking statements contained in the Private Securities Litigation Reform Act of 1995 as amended. Please see the last page of text in this morning's earnings release for additional information about the risk factors associated with these forward-looking statements.
If needed, a copy of the release is available on our website or by calling the investor relations department at 210-220-5234. At this time, I'll turn the call over to Phil.
Thanks, A.B. Good afternoon, everyone, and thanks for joining us. Today, I'll review third quarter results for Cullen/Frost, and our Chief Financial Officer, Jerry Salinas, will also provide additional comments before we open it up to your questions. In the third quarter, Cullen/Frost earned $106.3 million or $1.65 per share, compared with earnings of $95.1 million or $1.50 per share reported in the same quarter of last year. This compared with $116.4 million or $1.80 per share in the second quarter. We continue to focus on our organic strategy while the economy works to move past supply chain issues and other lingering effects of the pandemic.
Average deposits continued their strong increase in the third quarter and were $39.1 billion, an increase of 19% compared with $32.9 billion in the third quarter of last year. Overall, average loans in the third quarter was $16.2 billion, compared with $18.1 billion in the third quarter of 2020, but this included the impact of PPP loans. Excluding PPP loans, third quarter average loans of $14.8 billion were essentially flat from a year ago, but up an annualized 6% on a linked quarter basis. Looking forward, we're very encouraged about the outlook for loan growth. New loan commitments booked through the third quarter, excluding PPP loans, were up by 11% compared to the first nine months of last year. For the quarter, new loan commitments were up by 6% on a linked quarter basis.
We were especially pleased that the linked quarter increase was due primarily to C&I commitments, which were up 30%. Our current weighted pipeline is 41% higher than one year ago and 22% higher than last quarter. The increases are in both C&I, up 22%, and CRE, up 28%. The market continues to be very competitive. In the third quarter, 69% of the deals we lost were due to structure, compared to 50% in the quarter before. We also continue to add to our commercial customer base, and we recorded 619 new commercial relationships during the quarter. While this was down from the same quarter a year ago when we were experiencing incredible PPP success, it is two-thirds higher than the quarter immediately before the PPP program.
As with the second quarter, we did not report a credit loss expense in the third quarter. Our asset quality outlook is stable, and in general, problem assets are declining in number. New problems have dropped to pre-pandemic levels. Net charge-offs for the third quarter totaled $2.1 million, compared with $1.6 million in the second quarter. Annualized net charge-offs for the third quarter were five basis points of average loans. Non-accrual loans were $57.1 million at the end of the third quarter, a slight decrease from the $57.3 million at the end of the second quarter. Overall delinquencies for accruing loans at the end of the third quarter were $95.3 million, or 60 basis points of performing loans. These are manageable pre-pandemic levels.
What started out as $2.2 billion in 90-day deferrals granted to borrowers early in the pandemic were completely gone as of the end of the third quarter. Total problem loans, which we define as risk grade 10 and higher, were down slightly to $635 million at the end of the third quarter, compared with $666 million at the end of the second quarter. In the third quarter, we continued making progress toward our goal of mid-single digit concentration level in the energy portfolio over time, with energy loans falling to 6.5% of our non-PPP portfolio at the end of the quarter. Our teams continue to analyze the non-energy portfolio segments that we considered the most at risk from pandemic impacts. As of the third quarter, those segments are represented by restaurants, hotels, and entertainment and sports.
The total of these portfolio segments, excluding PPP loans, represented $695 million at the end of the third quarter, and our loan loss reserve for these segments was 8.8%. The credit quality of individual credits in these segments is currently mostly stable or better compared to the end of the second quarter. We reported in the second quarter that we had completed our 25-branch Houston expansion initiative, and we're very pleased with the results. We've identified eight more locations to open in the coming months, and that process is underway. Let me update you where we stand through the third quarter with the Houston expansion, excluding PPP loans. Our numbers of new households were 134% of target and represented more than 12,200 new individuals and businesses.
Our loan volumes were 177% of target and represented $371.4 million in outstandings. About 80% of this represents commercial credits, with about 20% consumer. Deposits surpassed $500 million and were 111% of target. Commercial deposits accounted for two-thirds of the total. In the meantime, we're also preparing for our upcoming 28 branch expansion project in the Dallas region, which will kick off with the first new financial center opening early next year and continuing into 2024. The Dallas expansion will follow our Houston model, and we will employ the lessons learned during our team's successful rollout. I'll continue to emphasize that the business we are generating through our expansion strategy is consistent with the overall company.
Its profitability is weighted towards small and mid-size businesses, but it also is complemented by wealth management, insurance, and of course, consumer banking, which continues to see tremendous growth. For example, through the first six months of this year, we had already surpassed consumer banking's all-time annual growth for new customer relationships, which was 12,700 in 2019. At the end of the third quarter of this year, this had risen to 19,974 net new checking customers. That's already more than 150% of our previous annual record. We've worked hard to lower barriers to entry for potential customers with improved product offerings and physical distribution.
For example, besides Overdraft Grace, which we introduced in April, and Early Payday, which we announced in July, we also recently established an ATM branding partnership with Cardtronics that resulted in us having, by far, the largest ATM network in the state. In addition, after over two years of study, we've begun the process of putting in place the infrastructure to add residential mortgages to our suite of consumer real estate products in late 2022. This will complement our portfolio currently consisting of home equity, HELOC, home improvement, and purchase money second loans, which has steadily grown to in excess of $1.3 billion. Utilizing best-in-class technology will allow us to provide Frost level of world-class customer service as we build this portfolio over time in response to customer demand. Finally, I wanna commend our team working on PPP loans.
Nearly 90% of the 32,500 loans for $4.7 billion have already been helped through the loan forgiveness process. That includes upwards of 97% of the first-round loans from 2020. Our team continues to put in outstanding work to execute our strategies, whether it's PPP, our expansion projects, or the enhancements we've made to our customer experience. I'd like to thank everyone at Frost. I believe we've got the best team in the financial services industry. They are why I continue to be optimistic about our company and our prospects for success. Now, I'll turn the call over to our Chief Financial Officer, Jerry Salinas, for some additional comments.
Thank you, Phil. Looking first at our net interest margin. Our net interest margin percentage for the third quarter was 2.47%, down 18 basis points from 2.65% reported last quarter. The decrease was primarily the result of a higher proportion of earning assets being invested in lower-yielding balances at the Fed in the third quarter as compared to the second quarter, and to a lesser extent, the impact of a lower PPP loan volumes and their related yields compared to the prior quarter. Interest-bearing deposits at the Fed averaged $15.3 billion, or about 35% of our average earning assets in the third quarter. Up from $13.3 billion or 31% of average earning assets in the prior quarter.
Excluding the impact of PPP loans, our net interest margin percentage would have been 2.27% in the third quarter, down from an adjusted 2.37% for the second quarter, with all of the decrease resulting from the higher level of balances at the Fed in the third quarter. The taxable equivalent loan yield for the third quarter was 4.16%, down 12 basis points from the previous quarter. Excluding the impact of PPP loans, the taxable equivalent loan yield would have been 3.74%, down six basis points from the prior quarter. To add some additional color on our PPP loans, total PPP loans at the end of September were $828 million, down from $1.9 billion at the end of June.
Forgiveness payments received during the third quarter were higher than we had projected, resulting in an acceleration in the recognition of the net deferred fees during the quarter. At the end of the third quarter, we had only about $11.5 million in net deferred fees remaining to be recognized, and we currently expect about 75% of that to be recognized in the fourth quarter. Looking at our investment portfolio, the total investment portfolio averaged $12.5 billion during the third quarter, up about $209 million from the second quarter. The taxable equivalent yield on the investment portfolio was 3.35% in the third quarter, down 1 basis point from the second quarter. The yield on the taxable portfolio, which averaged $4.1 billion, was 2.03%, up two basis points from the second quarter.
Our municipal portfolio averaged about $8.4 billion during the third quarter, up $230 million from the second quarter, with a taxable equivalent yield of 4.04%, down five basis points from the prior quarter. At the end of the third quarter, 78% of the municipal portfolio was pre-refunded or PSF insured. The duration of the investment portfolio at the end of the third quarter was 4.5 years, up slightly from 4.4 years in the second quarter. We made investment purchases towards the end of September of approximately $1.5 billion, consisting of about $900 million in MBS agency securities with a yield of about 2%, about $500 million in treasuries yielding 1.07% with the remainder in municipal securities.
Regarding non-interest expense, looking at the full year 2021, we currently expect an annual expense growth rate of around 3% over our 2020 total reported non-interest expenses, which is consistent with our previous guidance. Regarding the estimates for full year 2021 earnings, given our third quarter results and the recognition of lower PPP fee accretion for the fourth quarter, we currently believe that the current mean of analyst estimates of $6.48 is reasonable. With that, I'll turn the call back over to Phil for questions.
Thank you, Jerry. Okay, we'll open it up for questions now.
Thank you. At this time, we'll be conducting a question-and-answer session. If you'd like to ask a question, please press star one on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star two if you'd like to remove your question from the queue. For participants using speaker equipment, it may be necessary to pick up your handset before pressing the star keys. One moment, please, while we poll for questions. Our first question comes from Brady Gailey with KBW. Please proceed with your question.
Hey, thanks. Good afternoon, guys.
Hey, Brady.
Hey, Brady.
Well, I just wanted to start with mortgage. You know, I remember you guys got out of the mortgage business, I think it's been a couple decades ago, and now you're getting back in. Maybe just two things. You know, is this an originate and sell model, or is this an originate and keep it at cross model? You know, if it's originate and sell, how big of an investment do you think you're gonna make there? Like, will this be a meaningful fee income contributor?
Brady, actually, this is gonna be a. It will not be a sell model. We're wanting to do this to expand relationships, and it's our intention to keep all these loans on the balance sheet. It won't be really a fee-based model, it's gonna be more of a portfolio model. We'll be in charge of the experience from beginning to end, including servicing. As I said, it's our objective to make this a world-class customer experience that people have with, you know, with our general products. In addition, which is what we have with our current consumer real estate products that I mentioned in my comments earlier, which include too, you know, home equity line of credit, purchase money seconds and home improvement.
I'm really excited about this opportunity, and I think it's important, because I realize we were the poster child for getting out of the mortgage business before the meltdown. I think it's important to realize what we're doing and what we're not doing. What we're not doing is going back into essentially the model that we had utilized before, which was very consistent with what was out there, which was a commission-based sales structure, and one where technology, size was so important in order to achieve scale with the technology at that time. What we're doing today is this will be Frost bankers providing this service, and really it's in response to customer demand.
It'll be done through our branch network, which as you know, is expanding significantly these days into major significant markets in the state. I think really importantly, this is a differentiator between where we were 20 years ago. The technology that's available today is much more accessible. It's much more scalable at an appropriate level. There's a lower cost to entry and a lower cost to support as we use cloud and SaaS technology providers. It's mainly about providing a great customer experience in this sector, and I believe we've got a great opportunity.
Phil, how big would you like resi mortgage to be over time, like as a percent of total loans at Frost?
You know, our consumer real estate today runs about $1.3 billion, which, so it's a little below 10% of the portfolio. You know, I think pound for pound, residential mortgage could be, you know, at least that size in the next five years. I would think it has a chance to be bigger than that. That won't take a tremendous amount of volume, frankly. I think if we got, you know, a little bit less than two loans per month for our branches, that would be a good target for us to shoot for. In the markets that we're in, that'll give us some pretty good volume over time.
All right. Then my second question is just more about deploying cash into the bond book. I mean, the pile of cash you guys are sitting on just keeps growing. It's now 35% of average earning assets. You know, we've seen the long end of the curve move a little bit from 90 days ago. I mean, the 10-year bond yield's now at almost 1.60. How do you think about the timing on, you know, when you start putting more of this cash to use in the bond portfolio?
Well, Brady, just for some perspective. Right now, really as we look out into the fourth quarter, I only see projections for, say, another $750 million being spent this year. We did $1.5 billion in September. I see us maybe doing another $750 million . You know, we're not gonna get rushed into it. Obviously, you know, we spend a lot of time looking at what's out in the marketplace, looking at those yields, just as you've discussed. We really like the optionality that we have. We're not gonna be rushed into something. You know, I don't think we've got a timeframe in mind to do that. Obviously, if the yields turn our way, we'll jump in.
We're not in a rush to do that. I think that we feel like when the right time comes, we'll jump in. We're not afraid to make purchases. We just don't feel like it's the right time right now given the current environment.
Okay. All right, great. Thanks, guys.
Sure.
Our next question comes from the line of Ebrahim Poonawala with Bank of America. Please proceed with your question.
Good afternoon.
Hey, Ebrahim.
Just one follow-up, Jerry, on the PPP fees. Sorry if I missed it. What was the PPP fee in the third quarter?
The fee itself, or you want all the interest income?
The fee-
The total interest income associated with PPP loans was $30 million.
$30 million.
That was down from $45.5 million in the second quarter.
$45.5 million. You mentioned that there's $11.5 million to go, plus whatever the 1% on the rest of the loan balance is.
Correct.
Understood. If I x out the $30 million, NII would be about whatever, for the third quarter, around $240 million. Is that a good base? Like, do you think NII should grow from there?
Help me, Ebrahim. Are you talking about net interest income or?
Yeah, net interest income. When you look at it from an ex PPP basis, has that kicked off?
Yeah, let me get that. Let me kind of look at my information here. Yeah, I think on a non-PPP basis in the third quarter. I think that if you take that, I'd say that really, again, with the purchases we made late in the third quarter. You know, you need to take that into account. Yeah, that run rate I would say is really kind of a bottom, if you will. We'll get a little bit of that PPP in the fourth quarter. If you pull that out, I think from here on, I mean, you heard Phil Green talk about the loan growth that we've had in the third quarter.
The pipeline looks good, and then I mentioned that we made some purchases in September that we really didn't feel the impact of it in the third quarter. Yeah, from the way I look at it, hopefully, that was the bottom as far as net interest income dollars.
That's helpful. I guess just going back to the mortgage conversation earlier, Phil. You announced a partnership with Blend, I think, in middle of October. Just talk to us, one, is there more to it? Beyond mortgage in terms of what you think that partnership does for you on the consumer side. What else are you looking at? Like, you've been a little bit ahead of the pack in terms of either reacting or staying ahead on overdraft fees, on early payments for employment checks. Just give us a sense of what we should make of the Blend partnership and what else are you looking at when you think about fintech partnerships?
Well, Blend has been a product that we have been utilizing, you know, I know it's been well over a year, it was pre-COVID. This is an expansion of that. I'd say also, Ebrahim, you know, as we look at the other segments, you know, you've got the application and the decisioning, and then the servicing piece. We're looking at best-in-class software for all three of those segments, and it could be that, you know, those last two could be one provider, it could be two providers. It's exciting because it's not often in this business that you get to really choose the technology you want for a strategy and where you don't have a legacy system involved that you've got to work off of.
It gives us the ability, along with the work we're doing with Infosys to create the best workflow and the best customer experience with the best software. What I think we're gonna see happen is it will take the software that we've utilized for the mortgage experience and the mortgage product, and we'll see that reengineered into our workflow that we currently do for that large portfolio of consumer real estate. I think there's some real advantages that we can look forward to getting in this.
Understood. Just one more, if I could. Like, you sounded pretty upbeat on the outlook for loan growth. Give us a sense of, is that growth based on optimism that things start easing on the supply chain and customer sentiment improves? Or are you actually seeing tangible signs of activity levels picking up where you do feel like fourth quarter and at least first half of next year should be pretty good loan growth-wise?
You know, in talking with the people, they’re optimistic. I mean, a couple of additional things. You know, our look-to-book ratio has been, I think, pretty good versus a year ago. Our, you know, our numbers on looks up 33%, and our bookings are up 33%. So it’s good to see that those are consistent with each other. You know, one thing to keep in mind is as we continue to adjust the energy portfolio, and you look at that linked quarter, you know, increase, which was 6%, not annualized for new commitments booked. That included a 58% reduction in energy commitments booked. Still a little headway there as we rationalize that portfolio where we wanna be, you know, still in the mid-single digits, still an important portfolio.
As you know, we've been working down our concentration there. Then I think another thing is that I look at is if you look at our and I mentioned this, our pipeline, essentially, let's call it ninety-day outlook. The weighted pipeline, as I said, it was about 41% over last year, 22% up on a linked quarter basis. That weighted pipeline is, you know, what our people expect to happen. That tells me that they're somewhat optimistic on what we're gonna actually be able to execute there. I'll say another thing, you know, as I've looked at just utilization under lines of credit, we've seen that pick up, particularly in the commercial business. I think it hit a low point in May.
Utilization working capital lines, this is for C&I, was about 28.7%. At the end of the third quarter, I think that was 31%. Pretty much 31% even. We've seen that contribute to some volumes. I think we'll because we've had some good execution on commercial real estate recently. You know, we're sort of early in the game on those projects. As they mature, I think we'll see some funding up on that. You know, just the sense we get is that's really indicative of what we've been experiencing over the last several months, as opposed to, you know, people expecting supply chains to get better, that type of thing.
Because frankly, I don't think the supply chains are getting any better, and labor is not any better. To see this in that environment was encouraging.
That's helpful color. Thank you for taking my questions.
Mm-hmm. Thank you.
Our next question comes from Jennifer Demba with Truist Securities. Please proceed with your question.
Thank you. Good afternoon. You said you're gonna kind of employ some lessons learned from the Houston expansion with the future branch adds in Dallas and the rest of them in Houston. Can you just talk about what you have learned over the past two or three years with this branch expansion?
Well, don't wanna give away all the secret sauce. How about this? It's all about people, right? It's all about getting the right people and getting them early enough, but before you open up these locations to get them engaged with the market and make sure that you're bringing in people that really believe in the culture of the company and what we're trying to do.
As I've said before, we haven't done and don't do teams of people. We do individuals who are making a decision that Frost is the right place for them to be. I think it's really the success of our Houston strategy was really related to having great focus, and commitment of resources on our part of really talented individuals to lead that effort, and then selecting the right people that have really executed to create these results. That's what it's gotta be. You know, I will say it's a little, probably a little tougher market than it was when we started Houston three years ago, 'cause some of the supply chain stuff, some of the cost stuff, you know, people are not getting any easier to hire.
I think we've had pretty good results thus far. I think we can and we are utilizing the same procedures in the company and the same focus that we did in Houston. Let's just say that. I'm very confident of our ability to execute that plan.
Can you just talk about, you know, someone asked before about the excess liquidity. How long do you think it's gonna take to deploy Cullen's excess liquidity? You guys have a lot more than most of your peers.
You know.
Yeah, what I'd say, Jennifer, is number one, you know, we typically have more liquidity than most. Obviously, you know, we're not gonna carry 35% of our earning assets in cash. Let me give that sort of backdrop that we do tend to keep a little bit higher level. I'll say that we won't jump in all at once. I think that it'll take, you know, most of 2022 to make a major dent in it.
Yeah, you know, Jennifer, you know, with the loan-to-deposit ratio where it is, hovering around 40%, that's kinda where we were at the last low point historically for our company. You know, when you look at what it took to get to, let's say an 80%, you know, high 70s-80%, I mean, that was an extended period of time. I think whenever we saw the last decline in loan-to-deposit ratio, it was into the 50s, and that was in the Great Recession. I think it took us five years to get back to a, you know, for us, a more normalized number. I mean, honestly, I think it's gonna be at least that, for us.
Because one thing is, as I've always said, that this is a high-class problem, because if you're doing your job growing deposits, you've got plenty of liquidity and plenty of dry powder, and you're able to generate funding at a really good competitive cost because you can generate so much demand deposits, which are, you know, very low cost. You know, I'm not so much worried about utilization of liquidity per se, as I am just about making sure that our loan growth is solid, our new relationship growth is solid, and our strategies for organic growth are being successful.
I think the you know the earnings will take care of itself, and then we'll be opportunistic, as you've seen us be, on when we employ that liquidity into the bond market, or frankly, if we don't employ it into the bond market. I mean, frankly you know the optionality that Jerry talks about, I think becomes increasingly important as we see what happens with inflation. Not just supply chain, but what we're seeing with labor costs. You know, I tend to think that the short end is gonna have more volatility than the long end over the next year or so. Just my opinion. But we've got the opportunity to take a look and see on that.
Thank you.
Mm-hmm.
Our next question comes from the line of Steven Alexopoulos with J.P. Morgan. Please proceed with your question.
Hi, good afternoon. This is Anthony Elian on for Steve. Phil, you mentioned that you've had close to 20,000 net new checking accounts added so far this year. Are you able to parse out how much of these new customers are choosing Frost because of services you offer, including Early Payday and Overdraft Grace that not many banks offer?
You know, I can't tell you that because they don't tell us, you know, why they're joining us. I can tell you, though, that, you know, Early Payday was, let's see, July. The overdraft grace in April. You know, we broke the record for our all-time annual number in June, so you'd have basically seen no impact there for Early Payday and pretty limited impact from overdraft grace. I think one of the things that's really helping us, you know, aside from just reputationally what, you know, our service Net Promoter Score tells us about how people think about us is, I mean, the Houston expansion strategy has been a really big contributor to growth. So there's that.
They are really helping us see increases there. Secondly, we've done a great job of developing our capability and marketing for online. I think 41% of the accounts that we've opened this year have been opened online. Doing a good job in both those areas in the digital side and then expanding on the physical side and the geographic distribution, both of those things I think have really helped us get to a new level in terms of consumer growth.
Okay, great. My follow-up, you mentioned in the prepared remarks you plan to open 8 more Houston locations in the coming months. Any more details you can provide on this in terms of, the expense run rate beyond 4 as you build these branches? Thank you.
You know, what I'd say to that is I think we've kind of given a little bit of this guidance before. We do have our planned Dallas expansion that Phil talked about, and then also our expanding on the locations in Houston that we've already opened. The guidance that we gave originally when we announced the Houston plan was that it was gonna cost us about $0.19 in the first year. That's really, you know, based on what I'm looking at, I think you'd be pretty close to that right now, given what we're looking at. Obviously, some of that'll be dependent on, you know, the timing of the locations, you know, the hiring of the people, et cetera.
I think you'd be pretty close, you know, about something, $0.19 in 2022.
Thank you.
Also, I guess I'll mention, you know, I think your comment mentioned that we do Houston all in a year, and those will come over a two-year period, and Dallas will be a two-and-a-half-year period. Those will be coming in over a period of time.
Our next question comes from Jon Arfstrom with RBC Capital Markets. Please proceed with your question.
Hey, thanks. Good afternoon.
Hey, Jon.
Hey, Jon.
Hey. Phil, one of your comments earlier about the competitive environment and 69% loss to structure versus 50 in the prior quarter, what's changed? Do you expect competition to get even tougher from here?
You know, I was talking to our people last week about what they're seeing. I'd say largely the same thing, you know, burn down rates, guarantees, terms, loan to cost, et cetera. The one thing that was a little different that they've seen on the permanent side was a trend towards, again, these are permanent deals, interest only deals for five and 10 years, which is sort of new. You know, haven't seen as much of it. I think that, you know, around the margins, around the edges, it continues to be, you know, more competitive.
You know, there was one. I remember they told me about one person who was putting together a deal. I guess they were a broker for a permanent deal for a customer, and they had said, "I don't know what else I can ask for. They said yes to everything." So I mean, that just tells you how accommodative the you know, the markets are being. All that said, though, our people are optimistic, you know, about you know, our ability to get deals and you know, we have great relationships with key customers. You know, you'd think that there'd be more pessimism as you see this thing get more competitive. I think just you know.
We'll, you know, we know what to compete on when, and yet still keep our disciplines. We seem to be doing okay.
Yep. Yeah, the new commercial relationship numbers are certainly good, and same with consumer as well. I guess that's, you know, that's what you control, right, is activity.
Exactly.
I guess back on the loan to deposit question, I know it's not the most critical ratio out there, but it seems to me that maybe the excessive deposit growth is starting to abate a bit, and you're seeing a bottoming in loans and, you know, perhaps over the next couple of quarters, that starts to reverse itself and climb back up again. Is that a fair way to think about it?
I think that certainly, deposit growth was a little bit softer in the third quarter. You know, it was still, you know, 9% on an annualized basis, so.
Yep.
You know, it's still strong, but it's not the double digits that we've been seeing, you know, the high teens. From that standpoint, you're right. I mean, if anything, at that level with loan growth continuing or starting to improve now, that number could get better. At the end of the day, you know, we're all about relationships as Phil said, and you know, if those deposits keep coming in, we want them, right? To the extent that those deposits picked up again, you know, we'd certainly be happy about that. I think theoretically you're right. If deposit growth slows down and loan picks up, yeah, that'll kind of take care of itself. You know, we're still adding new locations.
We continue to offer top quality service. We're still interested in opening new accounts. You know, when I look at a 12-month rollback, looking backwards about where our growth came from, in a rolling 12 months on, you know, the deposit side, you know, 70% of it was augmentation from our existing customers, and 30% of it was from new customers. That's a little bit higher than we've been seeing. We've been seeing like a 75/25. It's good to see that. Yeah, I think you're right, but we'd love to continue to see deposits grow.
Okay. Yeah, that was one of my next questions. I guess that 70-30 split I guess I was curious on that. I guess the last one, Phil, you touched on it in your very early comments about Houston when you were giving some of the numbers. You touched on the fee businesses. Can you talk a little bit about what you're seeing in, you know, from the new branches in terms of the fees and fee generation?
Jon, honestly, I don't have that breakdown with me in terms of that for the expansion branches. I can tell you on our last Jerry Talk a little bit about fee income. We've really had a nice year at this point in wealth management and wealth advisors. A little bit tougher in the insurance, but still, you know, still holding our own there. Jerry, you wanna talk a little bit about the non-interest?
Yeah, I think Phil mentioned, you know, we thought, I thought we had a pretty solid quarter on the non-interest income side. You know, the trust side of the business had some good growth there and you know, with the investment fees growing, I think 13%. Then good oil and gas revenues there. You know, obviously with the higher commodity prices, that's helping us compared to where we were, say, in the third quarter last year. They continue to have new accounts growth there. Good growth there. I think on the deposit service charges, you know, even with Overdraft Grace, you know, I think NSF OD combined, commercial and consumer, might've been down $100,000 compared to the third quarter a year ago.
We did see good growth on the commercial service charges with higher billable services. Our interchange fee I thought was really strong. We were up almost $1 million, and we had introduced a new commercial account that's responsible for about half of that. It's also interesting that, you know, we're seeing, you know, the numbers of transactions, debit card usage is obviously going up significantly and the dollar size of the transactions that are running through are also up. I thought we had a good solid quarter all the way around. A lot of customer derivative activity also in the other categories. Yeah, I think overall it was a good quarter. A little bit softer on the insurance side, as Phil said.
We've been fighting a little bit of an uphill battle there on the net new customer business there.
Okay. All right. Thanks for all the help.
Sure.
Thank you.
Our next question comes from the line of Ken Zerbe with Morgan Stanley. Please proceed with your question.
All right. Great. Thank you. Just a couple of questions on expenses. I heard you guys say you still expect expenses to be up 3% for the year. I guess, what kind of range are you putting around that 3%? Because if I actually put in 3% exactly, it does imply a very sizable increase in fourth quarter expenses, like just over $230 million. I just wanna make sure I'm thinking about that right.
Yeah. No, I think it's a fair question, Ken. I was looking at it this morning, and the one thing I'd point you to is our historical growth between third and fourth quarters. You know, when you go back and look last year, it's not too different than that. We do have some incentives that some incentive awards that get awarded in the fourth quarter. Some of them, because of the nature of the award and the age of the recipient, get recognized immediately. That's gonna have those awards they'll happen in October, and so that's gonna affect salaries in the fourth quarter. And then also marketing. You know, historically, we've been strong in marketing expense in the fourth quarter, and I expect that we'll see that as well.
Yeah, I agree with you. The numbers do look high, but if you look at our history and look at where our projections are at, I think we're right in there. Would I love for it to be lower than that? Sure. But I think right now, based on what I'm seeing, that's really where we're at. I think those two things, it's really that salaries category and the other expense category where I'm seeing most of the pressure.
Got it. Understood. Okay, thanks. Just a sort of a follow-up back to 2022 expenses. I appreciate the $0.19, totally makes sense. But what is the base? I mean, should we assume, let's say that the 3% growth happens in 2021, and then we grow another 3% on top of that, and then we add $0.19? I'm just trying to make sure I know-
Sure.
what we're talking about.
Well, obviously, we're not giving any sort of 2022 guidance. You know, we're still in that planning process. There's a lot of work going on there. I will say that, you know, Phil alluded to a little bit about the inflation that we're seeing out there. We are starting to continue to see wage inflation. I was looking at something from the Dallas Fed this morning, if I can put my hands on it. It was just interesting, a couple of the comments that they were making about the Texas economy, and it said, lack of applicants still top hiring constraint, but pay demands are an increasing concern. Then there was another bullet point that said, average hourly wages now rising faster in Texas than the U.S.
You know, we are dealing with that environment, but we are in the planning stages, so don't hear me saying anything about what our guidance is for 2022 expenses. You're right. Yeah, whatever base level you believe you'd start off with on 2022, I'm suggesting that the Houston expansion 2.0, if you will, plus our Dallas expansion, is gonna cost us something around $0.19 over our more normalized run rate, if you will.
All right. Perfect. Thank you.
We've reached the end of the question and answer session. At this time, I'd like to turn the call over to Phil Green for closing comments.
Okay. Well, thanks everyone for your continued interest and your great questions. With that, we will be adjourned. Thank you.
This concludes today's conference. You may disconnect your lines at this time, and we thank you for your participation.