SouthState Bank Corp (SSB) 2020 Q3 法說會逐字稿

Good day, and welcome to the South State Corporation third quarter earnings call. (Operator Instructions) Please note this event is being recorded.

I would now like to turn the conference over to Will Matthews. Please go ahead.

Good morning, and welcome to South State's third quarter earnings call. This is Will Matthews, and joining me on this call are Robert Hill, John Corbett, Steve Young and Dan Bockhorst. The format for this call will be that we will provide prepared remarks, and we will then open it up for questions.

Yesterday evening, we issued a press release to announce earnings for Q3 of 2020. We've also posted presentation slides that we will refer to on today's call on our Investor Relations website.

Before we begin our remarks, I want to remind you that comments we make may include forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. Any such forward-looking statements we may make are subject to the safe harbor rules. Please review the forward-looking disclaimer and safe harbor language in the press release and presentation for more information about risks and uncertainties which may affect us.

Now I will turn the call over to Robert Hill, Executive Chairman.

Good morning, and thank you for being with us. I'm excited to be with you today and excited for you to hear from John and Will about the progress the team is making in building South State.

The Centerstate-South State partnership was about the long term, but you can clearly see the progress being made in the short term, progress with technology, products, efficiency, all of which have us uniquely positioned for the long term. But most importantly, it's about having a great team which we are fortunate to have and are continuing to build.

Our Board and our management team are united around the opportunities that are ahead of us, and our culture is growing in a healthy way that will only make us stronger in years to come. John Corbett's leadership along this journey has been excellent.

And I will now turn the call over to John for more insight into the quarter.

Thanks, Robert. Good morning, everyone. I hope you and your families are doing well. In light of the challenging environment, I couldn't be happier with the progress our team is making and the financial results that we're producing.

It was a solid quarter. Adjusting for merger costs, the company produced diluted earnings per share of $1.58 for a 17% return on tangible equity. Our pre-provision net revenue grew to $170 million for a pre-provision return on assets of 1.8%, and tangible book value per share grew at an annualized rate over 15%.

The highlight for the quarter was clearly the profitability of our mortgage and correspondent banking units. We heavily invested in these fee-based businesses when treasury rates fell after the 2016 Brexit vote. And now 4 years later, these noninterest income lines of businesses are producing a healthy return on investment.

2 weeks ago, we announced another investment into our correspondent banking division with the acquisition of a broker-dealer based in Memphis, Tennessee. Duncan-Williams is a 51-year old family firm led by the son of the founder, and it's the perfect complement to the fixed income line of business that we've been building for over a decade. South State currently serves nearly 700 community banks nationwide, and Duncan-Williams adds an additional 250 financial institutions to our coverage universe. This was a negotiated transaction that Steve Young and Brad Jones have been pursuing for 1.5 years, and we're very excited to welcome Duncan and his team to South State.

While fee income has been remarkably strong, we continue to be faced with industry-wide loan growth and margin headwinds as we navigate through the pandemic. The good news is that our loan pipeline bottomed out in August and is steadily rising every week. The pipeline is now 24% higher than the low in August, and it continues to grow as our clients become more confident in the future.

Asset quality metrics improved considerably during the quarter with loan deferrals declining to just 2% of loans. Of the 2% of loans currently on deferral, half of those are, in fact, making their interest payments. So really, only 1% of loans are in full principal and interest deferral. And this is the second quarter in a row that we recorded only 1 basis point in charge-offs.

As for the merger integration, it's proceeding on time and on budget. And the encouraging thing to me is that the merger integration isn't slowing our forward momentum in the rest of the bank. I can definitely feel the South State team leaning forward and eager to go on offense to prove what this new franchise is capable of.

One area where we continue to improve is the South State digital experience. This month, under the leadership of Renee Brooks, we rolled out our new website, and next month, we'll roll out a new mobile banking app that has been in the works for over a year. Our team is excited to offer a mobile app that rivals the largest banks in the country for customer functionality and convenience. To pay for these digital investments, we will be downsizing our branch network by 20 branches this quarter, which reduces our branch footprint from 305 branches currently to 285 by year-end, which also increases our average deposits per branch to over $100 million, and that's up from $40 million a decade ago.

We're also excited about the huge opportunities that are presenting themselves to recruit the very best relationship managers from the largest banks in the Southeast. South State now has the scale, technology and capital markets platform to be the logical alternative for middle-market bankers looking to make a change away from the turmoil of the largest banks.

So when I step back and I think about our positioning and the environment that we're operating in, there are things that we can control and there are things that we cannot control. We can't control the path or duration of the COVID virus, and we can't control the shape of the yield curve. What we can control is the $80 million of cost savings to be achieved through our merger and branch consolidation initiatives. We can also control our investments into the future, investments in technology, investments in our mortgage and correspondent banking teams that are not dependent on the interest spread. And finally, we can make investments in top-notch relationship managers in the best growth markets in the country.

So as I think about where this franchise is headed. I'm confident we're making the right strategic moves to create value for our owners, our team and our communities over the next decade.

And finally, along those lines, I'm happy to share the news that South State has created a new executive-level position of Director of Corporate Stewardship. LeDon Jones, a 21-year veteran of our company, has accepted the invitation to lead our company's diversity initiatives, our college recruiting and management training as well as our ESG, community development and employee assistance programs. Everyone at South State is excited for LeDon in his new role and eager to support his leadership as we create a company culture that will be a source of pride for all of us in the years ahead.

With that, I will turn the call over to Will to provide more color on the numbers.

Thanks, John. Our net interest margin was 3.22% on a taxable equivalent basis, down 2 basis points from Q2. Given the June merger closing date and associated purchase accounting marks and pre-closing securities sales on the legacy Centerstate side, the second quarter NIM is not entirely an apples-to-apples comparison nor is the combined business basis margin of 3.38% from Q2 as it includes the unmarked Centerstate balance sheet and income statement for the 68 days of the second quarter prior to closing. Our margin continues to be negatively impacted by the significant liquidity we're carrying, $4.4 billion average for the third quarter. If you were to reduce our balance sheet cash and Fed funds sold to $1 billion, reducing deposit funding accordingly, our NIM would be approximately 24 basis points higher.

Loan yields of 4.35% were up 10 basis points from Q2, reflecting a full quarter of the Centerstate loan portfolio in the company. Accretion was $22 million for the quarter, and core NIM excluding loan accretion was 2.95%. As noted on Page 6 of the release, we had some measurement period adjustments as we finalized purchase accounting marks, including a reduction in the loan discount of $29 million. While this improved capital, I'll remind you that it will reduce future accretion accordingly.

Our loan repricing mix is 55% fixed, 25% floating and 20% adjustable. Our total cost of deposits continues to improve, down to 20 basis points for the quarter. Our CDs are relatively short with 19% coming due in Q4, another 32% in the first half of '21 and 26% in the second half of '21.

On noninterest income, we had a record $115 million in quarterly noninterest income, led by our mortgage and correspondent banking capital markets. Our mortgage team has done a really outstanding job in the midst of a merger of 2 mortgage teams as well as a pandemic, with record volume of $1.57 billion, 60% of which was purchased, and strong margins, resulting in $48 million in revenue. Our correspondent banking division continues to show strong results with a $26 million quarter.

And on that note, I'd like to echo John's welcome to the Duncan-Williams team. We're not disclosing transaction terms due to the size of this acquisition, but we are excited about Duncan and his group joining Brad Jones and his team in helping us grow this business as well as the help it should provide in a very low interest rate environment. Steve has responsibility for these noninterest income businesses and is available to answer questions on them during the Q&A session.

On expenses, our NIE for the quarter was $237 million, including $22 million in merger-related expenses for an operating NIE of $215 million. Our efficiency ratio was 55.8% excluding the merger-related expenses. Our expenses for the quarter came in a little better than we expected in part because we have begun to realize some of the merger cost saves thus far a little faster than expected through normal employee turnover and some departure of employees who will not be retained as well as certain vendor savings. We expect that this cost save realization number will continue to increase each quarter with the bulk of the savings coming in 2021, particularly Q3, after system conversion.

Additionally, our expenses for a number of items were down in Q3 due to COVID. Business development, loan-related and ORE expenses, travel expenses and to some extent, health insurance costs were all down due to COVID, but we would expect many of these to normalize once we are beyond the health crisis. So the Q3 run rate for several areas is lower than we would expect in a normal non-COVID environment. On merger-related expenses, we have recognized approximately half of the estimated $205 million to date, some of which occurred on the Centerstate side pre closing.

Turning to credit. Our net charge-offs remained very low at $594,000 for the quarter or 1 basis point annualized. Ending NPAs were 33 basis points of assets, down 5 basis points from Q2 due to a combination of payoffs and upgrades.

Our provision for credit losses was $29.8 million for the quarter, $22.1 million of which was for the reserve for unfunded commitments liability. After running 2 separate legacy bank CECL models and combining the results in Q2, we consolidated onto 1 model in the third quarter. And this consolidation of the legacy South State loans onto a different model resulted in the increase in the reserve for unfunded commitments.

For economic assumptions, we use the Moody's baseline forecast. That forecast has the unemployment rate for the South Atlantic region holding at 8.2% for Q3 and Q4 of this year before starting to decrease in 2021 with a forecast of 7.4% at year-end '21 and 5.4% at year-end '22.

With a provision expense of almost $30 million and net charge-offs of less than $1 million, our reserve coverage excluding PPP loans grew to 211 basis points including the reserve for unfunded commitments or 192 basis points just including the reserve for funded loans. This brings the allowance to NPLs to just under 4x.

Slide 8 outlines our loss absorption capacity ratio, which ended the quarter at 258 basis points. As John said, our deferrals reduced significantly since our last update, dropping below 2% at October 23. Additionally, our full P&I deferrals were only 1% at that date as almost half of the deferrals are paying interest. For the effective tax rate, our return to profitability in Q3 after the impact of the double count provision and other merger expenses in Q2 caused our effective tax rate to decline in the third quarter to 19.6% from the second quarter's 22.6%.

Turning to capital. With good profitability and a flat though still somewhat inflated balance sheet, our capital ratios grew during the quarter. Our TCE ratio grew 27 basis points, ending at 7.83%. Our CET1 and total risk-based ratios grew by approximately 80 and 100 basis points, respectively, ending at 11.5% and 13.9%. Our ending tangible book value per share was just shy of $40 at $39.83, up $1.50 from Q2 and up $1.63 from the year-ago quarter.

I'll turn it back to you, John.

All right. As a reminder, we are conducting this call from different locations, so it's going to be helpful if you direct your questions to the person that you'd like to respond.

This concludes our prepared remarks, and I'd like to ask the operator to open the call for questions.

(Operator Instructions) Our first question comes from Michael Rose with Raymond James.

I wanted to just circle back to expense as well. You guys had obviously some really good revenue generation this quarter and the expenses were down. I understand some of it is COVID related. But can you just help us from a run rate perspective?

As the economy continues to reopen hopefully, what are some of the specific add backs we should think about?

And if you could just remind us how much of the cost saves -- I'm sorry if I missed it, how much of the cost saves so far you've actually realized?

Sure, Michael. We would -- on the latter part of the question, we would estimate that thus far, we have recognized on an annualized basis run rate about 12% to 13% of the $80 million cost saves, so call that $2.5 million per quarter.

In the quarter, like I said, we had -- we don't have people out there doing business development efforts. There's a lot less travel going on. There's not much real estate foreclosure activity, and those related expenses and professional expenses associated with that, those areas are all down. And I think for a lot of us, we're not going to the doctor like we -- as often as we had, so we're seeing a little bit of decline in health insurance.

And I'd say all of those items combined relative to a more normalized quarter might be $5 million or so in a quarter roughly. And then -- so that's why I just wanted to be clear that while we're pleased that NIE was down to $215 million, this is an unusual environment and didn't want to -- people to over-interpret that as they look forward.

That's very helpful. And then maybe just switching to the core margin, 2.95%. I know there's some moving parts in there, first full quarter post integration. As we think about going forward, I mean, are there other areas that you guys are working on to kind of optimize the balance sheet? And then how does that relate to the ability to prevent or limit NIM compression from here?

Sure, Michael. It's Steve. Let me just kind of talk about a little bit in the broader context around revenue. We talked about -- on Page 18, we have a slide in the deck that talks about our revenue composition over the past 4 quarters. And what you see in there is the revenue composition has changed. A year ago, we were about 77% NIM, 23% fees. Now we're 30% fees, 70% NIM. So just wanted to draw your attention to that.

Our reported margin, as Will mentioned, was 3.22%, our core margin at 2.95%. Just a couple of things on that. We are a core deposit-funded bank in great markets. Our checking accounts make up 54% of deposits, and our total cost of deposits this quarter was actually 20 basis points. So we continue to grow low-cost deposits on our retail, our small business and our treasury platform. But from a margin perspective, we don't have a lot of room to reduce costs, but we will go as low as we can without cutting at the core.

Let me kind of speak through the components of margin. First, the yield on the loan portfolio ex PPP and accretion this quarter was 4.16%. We're putting new loan production yields this quarter in at 3.53%. So that continues to be a headwind.

As it relates to Will's comment on the excess liquidity, so we have about $4.4 billion of average Fed funds sold. It's probably about $3 billion over target. It weighed on our margin about 24 basis points this quarter. If you look at our -- pre merger, our total investment portfolio was right about $4 billion or around 12% of assets. Prior to close, we sold $1 billion of the Centerstate portfolio because it was going to be marked to market, around 1.25%.

So as we think about the future of that excess liquidity, we invested about $0.5 billion this quarter to get it to around $3.5 billion. We're likely on a path to get it closer to $4 billion by the end of the year. And then medium term, our target for the investment portfolio is around 12% of assets. That will depend upon liquidity. It will depend on yield curve. But we also want to make sure that we're not investing a bunch of capital in the lowest rate environment we've seen. So we are cautious on that, particularly as the noninterest income businesses have been performing.

One last comment on the securities portfolio. It was at 1.63% this quarter and we're adding purchases somewhere around 1.25%. So that would be the core margin comment.

The only other comment was on accretion. I think we had $22.4 million this quarter. Of course, that was elevated. We'd expect that to decrease from here. We have a disclosure in the earnings release that shows there's about $110 million of discount left on the acquired loans.

So with that, why don't I turn it over to Will, if you have any other comments there.

Yes. I think I just would reiterate a couple of things to make sure that we emphasize them. One is just the reminder Steve gave about the marking of the Centerstate portfolio in Q2 and the impact we had. So you basically had a couple of billion dollar bond portfolio, half of which you sold and turned into cash at 10 basis points and half of which you had to mark down to a yield of about half of what it was earning before. So that's the impact.

But the second is probably more important, and that is just our long-term focus and trying to make sure the decisions we make are ones that we're going to like in the long term. And it would certainly be easy to boost earnings a bit if we took that excess cash of $3 billion and invested it to pick up 110, 120 basis points over where we are earning today at the Fed, but we want to be thoughtful and do that over time and not get aggressive and then regret that 9 months or a year down the road if rates have moved back up. So we'll be thoughtful about pulling that lever, although it does exist.

Okay. That's -- all that color is helpful. So I think the way to read that is maybe core margin ex PPP and ex accretion income, probably going to see some pressure here, but you guys are making some investments. The loan portfolio and the pipelines that you mentioned will continue to grow. So maybe we get to a point where NII actually troughs on a core basis ex PPP and accretion sometime next year and then starts to build from there. Is that the kind of the messaging and the way to think about it?

Michael, it's Steve. I think that's a fair way. I think the tailwind will be any of our investment purchases. The headwind is the loan book until it gets closer to par. And with elections and COVID and all those, the yield curve is going to move around a lot, but that's how we're thinking about it.

Our next question comes from Stephen Scouten with Piper Sandler.

So maybe just want to get some clarification on the expenses. I want to make sure I heard it correctly. On the branch reductions, it sounds as though that's incremental to the cost saves related to the deal, but more or less, it's going to get -- we won't see net savings due to investments in digital and other technology investments. Is that correct to kind of interpret that way?

Yes, Steve. I think -- and John can elaborate further, but our -- it's hard to separate out between the merger cost saves we have as well as the additional investments we're making in digital and the branch reduction. And so we -- our $80 million goal includes all 3 of those, both the reduction in expenses from the branch reductions as well as the additional investments we're making in improving our technology. And really, it's more of a reallocation of the branch rationalization into the digital spend would be the way I would probably describe it.

John, you may have some better comments.

I think you nailed it, Will. I don't have anything to add.

Perfect, perfect. And then how would we think about maybe, I don't know, a variable comp percentage on mortgage -- with mortgage being so elevated, how do we think about maybe how that's represented in salaries kind of in this quarter and the quarters to come? Assuming that does kind of trickle down maybe with MBA forecast next year, how we can think about the impact on salaries or maybe efficiency ratio that you think about in that business on a variable basis.

Yes. Well, why don't I start, and then Steve, maybe you can comment on the efficiency ratio side. So 2 components to compensation expense and mortgage obviously. One is the staff to get all of the loans through the system. And that -- obviously, the need for support staff in an environment like this is greater than it is in a lower volume environment.

With respect to the variable comp associated with the mortgages, accounting guidance dictates that you offset against the revenues. So it's FAS 91. So the cost of originating that loan, i.e., the commission, is a revenue offset and the gain on sale margin.

Now I will admit to you that not -- we've looked at all our peers. Not everyone does it the exact same way. But we do, do it that way, and that's the way we understand the accounting guidance from the time we spent in discussion with a number of accounting firms. So that's a component that probably lacks some comparability when you look at other companies -- some other companies, I should say.

Yes. And Steve, I just -- to your point, mortgage for the industry had a great quarter. We are really super proud of our group. The integration of those teams, [Tom, Brent and Steven's] leadership, they're just doing a great job of integrating those teams.

So the production was very large, but the gain on sale is very large. So if you think about efficiency ratios, I would expect that over time, that margin, which are record at every company right now, would move back toward 3% even though the production, as we think about these historical levels, may move back 20%, 25% over time.

Hopefully, that kind of helps guide you through. The efficiency obviously is really good right now because the margins are so large. The margins will come in, but our -- and our volume will come down a little bit.

Got it. Okay. And then just maybe last one for me as it pertains to growth. And maybe this is kind of a John question, but maybe also summing up. It looks like there was maybe a big migration from the acquired book into the non-acquired book. I'm just -- or maybe not. Maybe if you can comment on that, the big reduction in acquired noncredit impaired.

And then just kind of with the pipeline building, how you think about net growth in this environment, which obviously is a little tenuous still at best, I guess, I should say. So just kind of commentary there would be helpful.

Yes. I'll comment on the first question, and Will, chime in to clean up my accounting knowledge. But I think, Stephen, you talked about a decline in the acquired book and a rise in the non-acquired book.

I mean you remember how this works. The acquired book only runs off, you never add to it. So all of the Centerstate loans that came in under the fair market value accounting, they only decline. And all of the loans that are being generated by both South State and legacy Centerstate, all of that goes into non-acquired. So I think you'll continue to see that mix where one portfolio is going down, the other one is going to have -- see outsized growth because now it's got double the production.

But from a growth standpoint, we've had a pretty volatile 2 quarters here, so hard to be precise in forecasting. But let me see if I can unpack the components for you a little bit here. And I think it's important to separate the commercial portfolio from the residential portfolio, and I'll start with the residential.

If you looked at our residential and home equity book at the end of the second quarter, we saw a $150 million decline, so annualized down like 10%. Well, on the same token, we had record residential production of $1.6 billion. We had $150 million runoff, but yet we produced $1.6 billion. Well, the economics, as Steve mentioned, of this gain on sale being at 4%, it's just unprecedented.

So it really doesn't make sense for us from an ALCO perspective to put on 30-year fixed-rate loans on our books at 2.75%. The right thing to do from an ALCO standpoint, the right thing to do for our clients is to move those loans to the secondary market. So that has been a headwind, residential, to our loan portfolio, but it's not been a headwind to the income statement in totality.

On the commercial side, the way I'd have you think about that is recognize that a commercial loan pipeline is a 90-day pipeline. So loans you close in the third quarter are typically loans where they enter the pipeline in the second quarter. So the -- what was the pipeline in the second quarter? It was dead. So it's not -- it doesn't surprise us that total commercial portfolio would be down in the third quarter.

But having said that, within the commercial portfolio, we were encouraged that we were up. The C&I portfolio was up during the quarter, and owner-occupied commercial real estate was up in the quarter. So it's important to think about those components. But as you think forward about the pipeline, the pipeline is climbing now between $100 million and $200 million a week.

And let me step back and put it in perspective for you.

Pre COVID, at the end of the first quarter, the pipeline was $3.4 billion. We hit a low in August at $2.4 billion. Since that low in August, the pipeline is back up to $3 billion, so a 24% increase in the pipeline. And we're feeling good that that's going to translate into the fourth quarter and the first quarter of next year into increased production. I don't think it's going to be any kind of rapid growth in the loan portfolio, but I think we will turn the corner and start seeing some modest growth.

Our next question comes from Kevin Fitzsimmons with D.A. Davidson.

Just wondering on correspondent banking. Given the strength, the full quarter impact of that coming in and the business is doing well but also the acquisition that you guys announced, just curious what -- how sustainable you view that pace of revenues going forward. Just any kind of seasonal or cyclical forces we should be aware of. But also with the deal -- are there other bolt-on deals like this you guys are looking at in that area?

Sure. Thanks, Kevin. This is Steve.

One of the things I'd just remind is, I mean, that we really like the diversification of these fee income businesses. If you think about mortgage, correspondent, capital markets and wealth, none of those businesses make up over -- more than 10 -- or more than 12% of our total revenue. So we like the diversification within each line.

As it relates to the correspondent group, if you look at the trailing 12 months under Brad's leadership, the division has done about $105 million or so of revenue. If you look at the components of that, about $20 million of it is in fixed income, about $80 million in our capital markets product and about $5 million in payments. So as you think about the future, I would think that the record year of $80 million in our capital markets business, likely with loan volume in the industry being where it probably will be next year, we'd expect that to come off a little bit, call it, 20%, 25%, just like mortgage probably will.

But just to go into Duncan-Williams, as we talked about already, Duncan-Williams has about 250 financial institution clients. You put that with our 700, that's about 1,000 financial institutions. Last year, in 2019, Duncan reported about $27 million in revenue. And just from a modeling perspective, as we model the deal, we typically try to run these noncapital-intensive businesses around the 75% efficiency ratio business.

So hopefully, that's helpful in your modeling.

So as you think about the pluses and minuses going into next year, I think you're going to see some of the capital markets activity likely decline a little bit, but with the opportunity to increase relative to our fixed income opportunity with our own group as well as Duncan-Williams and the replacement of revenue there.

And then I think, long term, I think we're just excited to see the synergies between the 2 teams as we have products and services from both groups that eventually, we can cross-sell into it. Clearly, that's not going to happen in the next 6 months, but that's the long-term approach. So hopefully, that's helpful for you.

That's great, Steve. Appreciate that color.

Just one other broader question about reserve build. I appreciate all the detail, the level you've taken the reserve ratio to and then more broadly, when you look at loss absorption as you guys are describing it. So that being said, if net charge-offs are staying low as they are and we don't get any real big change in some of the economic indicators for your CECL model, should we assume reserve build is mostly in the rearview for you guys? Or is it if you -- we still think there are losses coming, it's just they're not coming yet, they maybe will be later next year, it still may make sense for you all to incrementally build in the next few quarters?

Yes. Kevin, I'll start. And if Dan or anybody else wants to jump in, they can certainly do so. But the thing about CECL, of course -- and it's interesting that we all implemented this new life of loan model, life of -- shortly before hitting this pandemic. But the theory behind it is that we reserve fully for the economic forecast -- the losses that would be driven by the economic forecast and loss drivers in your model over that forecast period. In which case, that theory would tell you that today, you've got every dollar in your reserve that you need, and that should be true of everyone who's adopted CECL, absent a change in that forecast.

So forecasting from there, if the economic forecasts don't worsen, then our reserves should be adequate. Now there's a lot of cloudy guidance to the crystal ball right now with what appears to be increases in COVID cases and who knows what that's going to do to some of the forecasts of unemployment and other loss drivers going forward. But sitting here today, the way the CECL model is supposed to work, we should be fine.

I mentioned in my comments the model change. So the -- given the timing of our closing, we had to do really a sum-of-the-parts methodology for June 30. You couldn't really convert over to a new model and go through the model validation process before then. And so when we did consolidate this quarter, that's what drove the majority of the provision expense. $22 million of it was based on the different methodology for the reserve for unfunded commitments. So absent that, the quarter's provision expense would have been $6 million or $7 million.

In terms of loss realization, Dan, you might want to comment on Kevin's question about where we think losses and the industry are heading in the future.

Yes. Just from a future loss perspective, the last 2 quarters have been very good from a charge-off perspective, and I don't anticipate any material change in that here in the fourth quarter. It's more first, second, third quarter, depending upon how the pandemic plays out and what impact that may have in the future on credit. But right now, NPAs, charge-offs, et cetera, credit looks strong.

Our next question comes from Catherine Mealor with KBW.

One follow-up on the asset quality. You mentioned in your slide deck that classifieds increase this quarter. Can you just give us a little bit of color around the categories that drove that increase this quarter?

Dan, you want to take that?

Yes. Dan Bockhorst, I'll take that question. The pandemic created economic headwinds that put a lot of loans on deferral in Q2 and Q3. And as good risk managers, we did a comprehensive review with the credit administrators and market presidents in August of all of our loans over $1 million that were either in high-risk categories or were on deferral. As a result of this review, we made changes to the risk rating grades so that we could ensure that we have the appropriate allowance and also acknowledge the impact that the economic headwinds have had on some of the borrowers.

Clearly, there is more stress in the hotel book across the entire industry. And so that's what's driving these numbers primarily, the makeup of the majority of the classified and criticized, recognizing the headwinds there. The severity of any loss is mitigated by the approximate 5% PCD mark on the legacy Centerstate Bank loans plus the overall 55% LTV in the hotel portfolio. From where we are today, if the economy continues this rate of recovery, we don't anticipate any material change in the level of criticized or classified assets in the immediate future quarters.

Great. And on the hotel book, any kind of update on what you're seeing at some of your properties in terms of occupancy rates? And maybe kind of the difference between what you saw this past quarter in your coastal properties versus your metro markets.

Yes. The hotel book is performing better than we anticipated. About 2/3 of the portfolio is in leisure segment, vacation areas, coastal waterways or destinations within driving distance of a large segment of the Southeast population.

And the summer was fairly good. Occupancy levels in those categories exceeded 60% and in some cases, were greater than 80%. So -- and you combine that with deferrals and the PPP funds. So with this improved performance, it allowed a lot of these borrowers to stabilize and build some liquidity as well as they have adjusted expenses to better operate in this environment.

The business segment, which makes up about 1/3, we've seen occupancy levels typically in the low to mid-50% range. Some might be a little bit lower. Those are the ones that are struggling a little bit more and have a little bit more headwinds.

Okay. Very helpful. And then maybe one other question on just big-picture capital thoughts. A lot of banks are starting to talk about buybacks. I don't think many banks of your size will start buybacks this year, but how are you thinking about what you're looking for to be able to start to think about reengaging in the buyback activity as we move into next year?

Catherine, it's John. Maybe I can comment. And Will, feel free to chime in.

A lot of uncertainty this summer. Plus in our situation, we are putting 2 large balance sheets together. Wanted to see where these capital ratios shook out. We did raise some sub debt in the second quarter of the year to bolster the capital position.

If you look at us, we look pretty good relative to our peers on CET1 and total risk-based capital. The one ratio that's a little bit lower is tangible common equity. I think it ended the quarter around 7.8%. If we keep profitability somewhere -- where it's at today, that should exceed 8% headed into 2021. So if credit is as benign in '21 as we're feeling like it might be now, having that extra capital, getting TCE back above 8%, I feel like it gives us some optionality to look at a buyback.

Right now, the dividend, I think, is yielding about 3%. We're paying back about 1/3 of the earnings. I feel pretty good where the dividend is, but I think that the growing capital base, capital formation is going to give us some options going into 2021.

Our next question comes from Brody Preston with Stephens.

I just wanted to circle back on the expenses real quick. So appreciate the $5 million or so in business development that's sort of not in the quarterly run rate right now. But -- and correct me if I'm wrong, theoretically, that should have also been somewhat missing from the 2Q sort of pro forma run rate of $225 million, just given the world was still locked down at that point or just coming out of a lockdown.

And so I guess, it's still really good cost saves. And you mentioned the $2.5 million or so from the SSB/CSFL sort of cost savings. And so I guess, was there anything else that you all sort of did that drove the larger sort of reduction in expenses this quarter?

So Brody, it's Will. I'll -- let me clean up a little bit just to make sure. So that $5 million I mentioned, that was more than just business development. It was -- included ORE, loan related, foreclosure related, all those kind of expenses and included professional fee reductions not -- all related to sort of activity levels being lighter.

And I look on the business development, and that's -- but this is Q2 versus Q3. And I think part of this is while -- Q2, you probably have some Q1 expenses, i.e. employee credit cards and stuff like that, that are -- those bills are received and paid in Q2. So that's probably a little bit of noise there.

In terms -- so the full $5 million was not business development, just to be clear on that. We -- I would say Q2 had probably a little higher expenses associated with our COVID reaction, just getting things geared up in terms of responding with our facilities and things like that. So that was a little bit of a -- I would think that would be a decline from Q2 to Q3.

But I just -- my comments were just to try to give you a little more clarity on -- it feels to me like $215 million is not a permanent run rate given the unusual environment in which we're operating. And if we move back to more normalized environment, we're going to be out calling on customers. Greg Lapointe and his team are going to be hitting the road and spending some money on the business development side. Foreclosure activity is going to come back in, in the industry to some extent. So we'll spend a little more money on that. So I just -- I kind of want to just make sure that, that idea was out there.

And Will, I'd just add to that. I mean we showed in the deck the combined business basis, the expenses there. And if you think about -- sometimes our fee revenue businesses move these expenses up and down on the variable side. Our revenue was up on the fee income by about, I think, $6 million, $6.5 million. A lot of that related to an MSR adjustment.

So if you think about the variability of the fee income, there really wasn't, from a commission perspective, much difference in quarter 2 and quarter 3. So the quarter 2 expenses would be -- on a combined basis, outside of our cost saves, would be pretty reflective of the run rate where we're at. So sometimes the fee income business has moved that around a little bit. But really, in the third quarter, it did not. So hopefully, that's helpful commentary, too.

Okay. All right. Understood. And then on PPP, I think you've mentioned that you have $2.4 billion still on the balance sheet. I wanted to get a sense for the timing of that. But then also, could you -- I don't know if you've looked at this at all, but do you have any sense for how much of those deposits are still sitting on the balance sheet?

I'll start by saying, Brody, that our -- with respect to PPP, our crystal ball is probably more cloudy than some of the other commentary I've heard folks give. But I'll tell you what we know.

We have about 20% of our loans in the forgiveness process. They've entered the forgiveness process. During the quarter, we recognized about $8.5 million of the net fee -- PPP fee, leaving us about $53 million, $53.3 million I think, remaining at the end of the quarter.

It is really hard to tell when that forgiveness process will really kick into gear or how quickly the SBA will respond, will there be another bill passed post election that may include some sort of forgiveness. But I would say our rough expectation with all those qualifiers is you're probably looking at a Q1 and Q2 concentration. And I don't know at this point whether it's more heavily in Q1 or Q2, but that would be our best guess today.

I don't have a figure for you on how much of those deposits are still in the balance sheet. That's a good question. But John or Steve may have a feel for that.

Yes. Brody, this is Steve. I would just point you -- we don't know the exact answer, but I mean it's pretty obvious when you look at the picture on Page 19 of the deck, which shows the deposits from the first quarter to the second quarter. And what you see is $4.2 billion of deposits growth there. And that's when all the PPP money was going out, and we're flat from there. So the way I would characterize it, the deposits from those companies, although we don't have a specific answer, really haven't sped out yet, and our deposits are flat after that big ramp in Q2.

Okay. Understood. I guess I'm just trying to think about the potential deployment of excess liquidity. And so you've got 4 -- $3 billion in sort of excess liquidity, like you said, on the balance sheet. But I guess just thinking about the deployment of that, either into loans or securities throughout 2021. I guess as those PPP deposits flow out, we'll -- I guess that would sort of flow out with it. Or do you -- I guess, do you estimate that you sort of have excess liquidity still on the balance sheet that can be deployed into earning assets beyond the PPP deposits?

Yes. Brody, I think the answer is we don't know. I think what our target -- if you looked at our companies as separate companies, we run our investment portfolio at 12% of assets. That's how we've traditionally done it. When we have a lot of excess liquidity in the financial crisis, we probably ran it closer to 15%.

But I think right now, we're in the wait-and-see mode. We need to build the securities book back up to that medium-term 12% number and then see where the landscape is and see how those PPP funds because I do think it's uncertain, and we just want to be thoughtful as we do it.

And I would say -- Steve, I would also just add to that, that while, certainly, a portion of the excess liquidity is PPP, it's also just the liquidity that the Fed has pumped into the economy. Some of those PPP loan fundings have been used to pay payroll and things like that. But I think we are -- there is just an excess amount of liquidity that's with us right now based on the actions the Fed has taken over and above the PPP.

Okay. Understood. And on Duncan-Williams, thanks for giving the $27 million in revenue last year. Just want to get a sense for if their business has, I guess, been negatively impacted this year as a result of COVID or how they've been faring year-to-date.

Yes. We won't give those public numbers, but just I'll talk about our fixed income business. Our fixed income business is up this year, and it's really primarily because there's just more excess liquidity sitting in the financial space.

So as you think about all the excess liquidity we're talking about, it needs to be invested at some point, so do our clients. And so I think even though financial institutions like ours have been pretty hesitant to go too fast on this, at some point next year, the year, after, you'll start seeing that deployed. So I think this is another good business to be in the hedging with because I think the fixed income revenue will probably be a little stronger than that. But it's probably too early to tell.

Okay. And then the $110 million or so in loan discounts, is that the total discount? Or is there something else beyond that?

That's the total.

Okay. And so just wanted to ask -- those $22.4 million on PAA loan accretion income this quarter, I guess just thinking about the quarterly run rate moving forward. Understood that it's supposed to step down, but what should the quarterly run rate on PAA look like perhaps for 4Q? And when do you sort of expect those loan discounts to be fully accreted into income?

Will, you want to take that?

Well, I was hoping you would, Steve. I'm chuckling, Brody. I don't -- I hope you understand why. If you've asked me 3 months ago, I would not have guessed $22 million. It's just a hard number to predict based upon when payoffs occur, paydowns occur and things like that. But it's clearly going to -- it should decline from here. And you model it over the weighted average life of that portfolio, but then the weighted average life ends up being -- in my experience over years of acquisitions, always ends up being shorter than what you had modeled.

But if I were to throw out a number for you, I would worry that I would be viewing that number with more precision than would be appropriate given the difficulty. But if you -- the best way to do is just model whatever you think the weighted average life would be for that acquired portfolio, and that would sort of guide you to a number. But I wish I could give you a better number, but I really don't feel comfortable doing so.

Okay. Do you have -- do you know the weighted average life off the top of your head?

I do not actually, I don't. But the type of lending we do, you're -- and again, this would be from origination date. The type of lending we do, you generally think about it being in that 3- to 5-year range, but...

Yes. Will, I'd say for the Centerstate book, prior to merger, it was around 3.3 years, was the average life. So I think that is probably a decent point.

Our next question comes from Christopher Marinac with Janney Montgomery Scott.

Thank you for the color on the problem assets on the prior calls. I just want to drill down to the kind of classified and criticized trends just to understand. Do you see a path where those loans get upgraded in the next couple of quarters? Or do you think it's going to be more sort of stagnant for a while, get more visibility on the recession, COVID, et cetera, and then those will migrate back later?

Dan?

Yes. This is Dan Bockhorst. I think a little bit of combination of both. I do think there's an opportunity for some of those classified loans to get upgraded sooner than later. And then the ones that are in the criticized category, probably a little bit maybe longer path of 6 months to 9 months to see those start to get upgraded as we get a little bit more visibility. But I don't anticipate those to also -- I don't anticipate those to migrate and get downgraded.

Got it. That's helpful. So again, the reserve build really kind of counts all that into effect now as of these downgrades?

Correct.

Got it. Okay. And then just a follow-up for whomever. On the PPP, what are you seeing on fraud? Is that an issue to worry about? Do you need to set aside reserves for that even if it's not material at this point?

The fact that it hasn't really been a conversation at any of our various risk meetings or whatnot would lead me to answer, Chris, that it's really not a factor for us, and I'll knock on some wood as I say that. But I think we've built some pretty good processes and had involvement of local teams, which is a strength of our company throughout that process. So hopefully, we would be less subject to that than others. I don't think we've seen much of that.

John or Steve, do you have -- and Dan, do you all have a better answer?

Will, this is John. I think I've asked that question. And so far, I'm receiving confidence back that they're not seeing trends of fraud. So it's still early, but that's what we're hearing now.

(Operator Instructions) Our next question comes from Jennifer Demba with Truist.

I think all the topics have really been covered, but I'll ask one more. You announced another 20 branches you're cutting. Can you just talk about your willingness to reduce more branches or corporate real estate that's not currently planned right now in order to offset revenue challenges, should there be any?

Yes. Jennifer, it's John. With our merger, there really wasn't branch overlap. So I think both companies in the past have acquired a lot of banks with branch overlap, and we've put branches together. But setting that aside, there is the secular trend. And I think you're seeing this history with both companies that each year, looking to rationalize the branch network.

In fact, if you go back over the last decade -- there's a slide, Page 20 of that deck. If you take the 2 companies, put them together, there were about 85 branches a decade ago. We've acquired 420 branches, but we've consolidated 212. So we've consolidated about half of everything that we purchased. So this has been an ongoing trend, and we think that, that will continue to be an ongoing trend.

You've heard other folks talk, Jennifer, about the COVID driving more and more digital adoption. Interesting, we opened all of our branches in the company this month. They've been closed for -- I think since March, so call it, 6 months that the offices have been closed. We opened it this month.

I've talked to some of the presidents and said, "Well, now that we're open, what's happened with the traffic inside?" They said remarkably slow, that customers have become accustomed to doing business in the drive-thru, doing business digitally, and they are -- the traffic has not picked up considerably in the lobbies now that we're open.

On the digital side, just some stats for you. Year-over-year, digital deposits are up 67%. A year ago, this is people taking a picture of their check on their telephone, we were doing about 15% of our deposits that way. Now it's 25%.

As far as actually opening new checking accounts online, that's up 170% year-over-year. It was 10% of our accounts a year ago. Now it's 27% of our accounts. And consumer loans opening up online is up 90% year-over-year. A year ago, it was about 10% of our consumer loans. Now it's 19%. So I think as we think about the future, we'll continue to evaluate the rotation of brick-and-mortar expense into digital expense.

On the other corporate real estate front, when we did the merger and we analyzed our operations center space, we've got 2 major operation centers, one in Charleston with a few hundred people and one in Winter Haven, Florida with a few hundred people. But in actuality, we have 17 total operations centers. So there's 15 smaller ones. Well, that's -- those support teams have been working from home for 6 months, and it's been working fine. So I think it's very likely that as we go through this efficiency project with the merger, that we may see a significant reduction in the number of those smaller operation centers.

So these are secular trends you're hearing from others. And definitely, there's a lever for us to continue to pull on a year-by-year basis.

One more question, John. Do you think your revenue producers are as productive working from home as they are working from the office?

Yes. Good question. I mean our 2 quarters here, our loan production is down. So you've got to say, "Well, is that because the revenue producers aren't as productive? Or is that because the economy got shut down." So I'd like to believe that there is productive -- I mean let me answer it a different way.

Go back to the PPP process. Okay? The economy was shut down there. They were working from home. We did 20,000 loans in a period of about -- I want to say it was like 3 weeks. So I think that's the case study, PPP, that the ability to be productive is there, at least in that kind of crisis moment.

But I think our relationship managers love to get in front of their clients. So I think that's slowly starting to open up. And in the long term, you're probably going to see a mix of in-person and also more digital contacts.

Good news is on the RM front, we're having a lot of success now recruiting. And there's a lot of turmoil in the biggest banks, so it's a new world. We're all trying to figure out how much time to spend in front of a Zoom call and how much time to spend in person.

Yes. Jennifer, I'd just add just anecdotally. On some of these fee lines of business, whether it be mortgage, fixed income, capital markets, they're all working remotely and having record production. So I think the ability to get in front of the clients is a little harder. But at the same time, I think we're all figuring out that you can do some of this more remotely. So I think there's pluses and minuses out of both of them.

This concludes our question-and-answer session. I would like to turn the conference back over to John Corbett for any closing remarks.

All right. Thanks a lot. These are great questions, and I hope we've been able to provide some clarity this morning.

We're going to be participating in the Piper Sandler conference in a couple of weeks. But in the meantime, if you have any questions as you update your models, please feel free to reach out to either Will or Steve.

Thanks for joining us this morning. And I hope you have a great day.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.