Good day, ladies and gentlemen, and welcome to the Bank of the Ozarks Second Quarter 2018 Earnings Conference Call. [Operator Instructions] As a reminder, this conference call is being recorded for replay purposes. .

It is now my pleasure to hand the conference over to Mr. Tim Hicks. Sir, you may begin. .

Good morning, I'm Tim Hicks, Chief Administrative Officer and Executive Director of Investor Relations for Bank of the Ozarks. .

Thank you for joining our call this morning and participating in our question-and-answer session. .

In today's Q&A discussion, we may make forward-looking statements about our expectations, estimates, and outlook for the future.

Please refer to our earnings release, management comments, and other public filings for more information on the various factors and risks that may cause actual results or outcomes to vary from those projected in or implied by such forward-looking statements. .

Joining me on the call to take your questions are George Gleason, Chairman and CEO; Greg McKinney, Chief Financial Officer and Chief Accounting Officer; and Tyler Vance, Chief Operating Officer and Chief Banking Officer. .

We're very pleased to report our excellent second quarter results, and we'll begin by opening up the lines for your questions. .

Let me ask our operator, Brian, to remind our listeners how to queue you in for questions. .

[Operator Instructions] And our first question will come from the line of Ken Zerbe with Morgan Stanley. .

Why don't we start off on the core spread, I'm kind of trying to narrow this down a little bit, but just I'd love your broader thoughts on this, you mentioned in the release that over time going forward, you do think that core spread should eventually move higher.

But when I think about the pieces and sort of think about sort of, let's call it, rough mid-60s asset beta or loan beta, but then this quarter's deposit beta, which interest-bearing is 84%, like, how do we end up in a place where core spread actually expands? Like, it seems that something would meaningfully need to change, I suspect it's on the deposit beta side, but I mean, essentially are you calling for core spread -- or the deposit beta to actually come down next quarter?.

1, yes, we hope to improve that deposit beta in the third quarter and the fourth quarter of this year; and #2 is we hope that we'll get a little more lift on the loan side, our nonpurchased loan yields improved to 18 basis points in the first quarter of the year I believe it was, is that correct, Tim?.

It is. .

And 14 basis points in the second quarter. And obviously, that core spread would have diminished less than half of what it did, if we'd gotten the same 18 basis points lift on the loan side. So we hope we'll see better results on both sides. .

Okay. And just to mix up the questions a little bit. In terms of the growth, one that, I guess, your comments in the release was that you're seeing competitors offer aggressive credit structures in pricing. I would love to get more detail on that.

Is it -- like, how broad is that? Is it -- is there any geographic concentrations on where you're seeing it?.

Well, we've seen a number of new competitors enter the CRE space over the last year, and that has been more or less broad-based.

I would not say it's in every market with the same intensity and vigor, but we've seen a clear increase in competition in a lot of markets on a lot of product types, with competitors competing both much more aggressively on bill structure and on pricing.

We have always said, and no one should be surprised at our actions in this regard, we've always said that asset quality is the most important in the asset quality pricing growth equation for us, and that growth is a tertiary consideration.

So the way we have approached it and will continue to approach is we're going to hold very firmly to our longstanding asset quality standards. Those are not negotiable.

We will get as aggressive on pricing as we can to be competitive while still achieving our minimum return on allocated capital for each loan and growth will be the variable that we'll adjust up or down, depending on how market conditions play out.

So we're quite content to have less growth unless we can get that growth on credit quality and pricing standards that makes sense to us, and I would hope that any thoughtful holder of our company's stock would applaud that approach, that credit quality is just not something that you want to play with or take chances on, and return on equity is not something you want to get below a minimum target level, and if growth suffers because you're being disciplined, then so be it.

.

And this will be my last question, but I guess just on the growth piece, the way you're describing it and I totally understand what you mean, and I'm glad you're being conservative on credit.

But you mentioned that you're quote, "less certain" that you can actually outgrow last year, but it seems that what you're describing, it's -- it should be not just less certain, it should be just like growth could potentially come in meaningfully lower than the growth last year, given the increase in aggressive competitors.

Is that a fair statement? Or is there some reason why growth should accelerate in the back half?.

Well, Ken, let me tell you this, if we knew -- if our crystal ball was perfect and we knew exactly what growth was going to be, we would tell you and everybody else on this call what our growth number was going to be, so there would be no uncertainty about it.

But we do live in a world that has uncertainty and variables and many customers and many competitors, and it's a dynamic landscape.

So we have not thrown in the towel or don't consider it impossible for us to beat last year's number, but it is a more competitive environment than we envisioned, when we gave that guidance earlier this year and last year, and that has created uncertainty about our ability to hit that number.

I think the more important reality that our shareholders ought to be focused on is whether we come in at 75% or 125% of last year's growth number. Even on the low side of that, we would be producing an organic growth rate that's twice probably what the industry is producing.

So whether we're growing at 2x or 3x the industry, we're still generating excellent growth and certainly with our stock trading where it is, you'd buy that growth at a deep discount relative to the competitors.

So I think there is an excessive focus on wow! are they going to be a little more than last year? A little less than last year? Or a lot more than last year, a lot less than last year? And the reality is that our growth is going to be at a very good metric versus the industry. We don't know what it is going to be today.

We'll know that in 6 more months, but we think it's going to be very good growth and we're obviously not getting rewarded for it. .

And our next question will come from the line of Catherine Mealor with KBW. .

One follow-up on the growth.

Outside of just the competition, would you say, George, that the flow of quality deals has changed as well? Or is it really just more on the competition side of the equation?.

I would say the flow of quality deals has changed. Costs of material for construction are going up. Costs of labor for construction are going up, construction period interest costs are going up.

So that makes it more challenging for our sponsors to find and pursue viable projects that are going to meet their returns on their equity investments in the projects in an environment where you're seeing very little increases in rental rates or sales rates.

Most property types that we're seeing in most markets around the country, your rental rates or your sales rates are not materially different up or down than where they were a year or so ago. So your cost to produce the asset is higher, the value of the asset is up a little bit or down a little bit from where it was a year ago.

So it is making it harder for sponsors to find deals that make sense and we've seen sponsors shelve deals that didn't make their target equity requirements and so forth.

So you have an interesting dichotomy here in that you've got an environment where there are probably fewer deals that are going to meet our very rigorous credit standards, and you've got a lot more competitors in the space chasing those deals.

Clearly, it was a more favorable environment for us a couple of years ago when we could find more good deals than we could close that met our standards and there was much less competition in the space, that dynamic has changed, and I think one of the great strengths of our company is our discipline and willingness to do what's right, even if it impinges on our growth numbers.

We would've had no problem achieving growth -- I think our growth through the first 6 months of the year was about 2% ahead of last year's growth rate through 6 months.

We would've had no problem achieving growth 20%, 30%, 40% ahead of last year's had we been willing to sacrifice our standards to do that, but that's not the way we run this company, and not the way that we have had better than industry credit metrics every year since we went public 21 years ago, and those credit losses over that 21-year period of time averaged 32% of the industry's average credit metric, and it's because we've been disciplined.

We're not perfect, but we've been much better than average because we've been very disciplined. .

That's very helpful. And then maybe one other one is just on the indirect RV and marine portfolio.

Can you just kind of give us some more detail and updated background on that portfolio, given it's now $1.5 billion and with most of the growth this quarter, just can you talk about some of the underwriting standards of that portfolio? The average size of credits? Kind of profiles, your typical customer, yield on that portfolio, those kind of things, and then how large you think that portfolio can get over time?.

Okay. Right. Great question. Thank you. One of the gems in the Community & Southern Bank acquisition that we did in July of 2016, and predominantly in Georgia was the indirect marine and RV business. They were in the indirect Marine, RV and auto loan business.

We looked at all of that and really did not like the auto loan business at all, felt like there was a lot of risk there and that it didn't yield a particularly high return by our standards. On the other hand, we really liked the Marine and RV business because they were pursuing a very high-quality customer, and were getting better yields for it.

So what I would tell you is in the recent quarters, the credit scores on our customers in that book averaged, Tim, 780 to 790. .

Originations this year have averaged 792. .

792 credit score on this year's originations. So it's a high prime, super prime, type customer base.

There's a strong emphasis on lifestyle customers, people who have owned multiple RVs and multiple boats before, so they understand what they're getting into and that combination of emphasis on high credit scores and lifestyle customers typically means that your customer base has high-income metrics and high balance sheet metrics.

So as we've said in the management comments, we have grown really excited about this unit because it is a unit that allows us to get much more exposure and diversification of our portfolio to the consumer sector while at the same time being true to our standards of high credit.

So many ways that you would get exposure in the consumer credit environment, you wouldn't get that high credit and you wouldn't insulate yourself from wide swings in credit performance and in economic cycles.

We feel like this business really gives us a high net worth, high income, high credit score customer that will be resilient in economic downturns, and is a good exposure and a good diversification of our portfolio toward the consumer sector.

The question about yields on the portfolio or yields on the portfolio net of the amortization of premiums that you pay for that paper, we do pay a premium for that paper.

So net of the amortization of that premium over the expected useful average life of those loans, the yields on the portfolio are just very, very close to the yields on our total portfolio of non-purchase loans so it's not dilutive or accretive in any meaningful way to our yields on nonpurchased loans, it's really tracking right in line with the average on that.

So it's a good yielding portfolio with high credit quality and RV. How big can it get? We think it has significant growth trajectory, as you can see on I think it was page 7 of our management comments.

There is a graph at the top of that, that shows the growth trajectory of that and the growth in the number of RV and marine dealers that we're doing business with across the country. We expect both the volume of the portfolio and the number of better relationships to continue to grow.

You noted that it was a lion's share of our nonpurchased loan growth in the quarter just ended, I would note and caution you to not expect Q2's growth to necessarily be repeated in Q3 and Q4 because it is somewhat of a seasonal business and in most years, we would expect that the second quarter based on seasonal factors would be the largest quarter of growth for the portfolio.

.

And our next question will come from the line of Timur Braziler with Wells Fargo Securities. .

My first question is around the deposit base.

Looking at betas versus growth this quarter, I would love to just hear any more color on what was driving the beta specifically in Q2? Was that all posted rates? Was that all negotiated rates, kind of any color behind that would be helpful?.

Tim, I don't know that I can give you a lot of meaningful color, I would tell you it was in part negotiated rates, part RFP rates.

We had some relationships that we had public funds and other relationships that were under RFPs from earlier dates that those cycles of that business came up to reprice that, rebid that, so we got a little push in our cost of funds from some of those.

And then it was in some markets on some product types posted rates as well, so it was really a combination and I don't have a breakdown and, Tyler, I doubt you have a breakdown at your fingertips. .

I don't have the breakdown. .

So that's about all we can tell you. It was kind of a combination of factors. .

Okay, and then going back to Ken's question and your answer about looking at different strategic initiatives currently to try and lower beta, going forward.

Any additional color you can provide on that? And what's the opportunity to really tap some of the more rural markets and align the deposit market share closer to the branch market share?.

Well, certainly, I think, there is a tremendous capacity to align our market share in line with our branch infrastructure over time as we need to do that. So that really is what I would say about the second part of that question.

The first part of that question, the answer is yes, I could give you a lot of details about actions that we're taking or initiating to try to reduce that deposit beta in Q3 and Q4, but I'm not going to do so for competitive reasons, if you don't mind.

I would rather be effective in doing what we're doing and implement that without telling all of our competitors exactly what we're doing as opposed to telling you about it, and then have trouble getting the results we want from it. So we do have a plan of action about how we're approaching that.

We were not pleased with our deposit betas in Q2 and we feel like we can refine some of our approaches and strategies in that regard, not changing them in any broad wholesale way, but just tweaking how we're doing that and get better results in the coming quarters.

And there is no guarantee about that, but we think we've got an approach that will be helpful. .

Okay. That's understandable. And then just 1 more for me, just circling back on the indirect RV and marine portfolio.

As you think longer term, obviously credit has been phenomenal for the bank as a whole, but as you think longer term for this product set going through the next cycle whenever it may be and how severe it may be, just can you give us some color around just broader allowance methodology? Allowance trajectory? Should we expect to see kind of continued build of allowance as that portfolio gets bigger? Or are you thinking the inherent loss profile there isn't going to be very different than what you've seen at the bank as a whole?.

1, those are longer duration loans on their contractual face amount, so we're holding more allowance for the longer duration of those, and 2, just consumer credit is going to have a slightly higher loss profile than our RESG portfolio even if it's high prime, super prime type credit, we believe, based on our modeling.

So it's a relatively modest difference, but not concerning to us at all. And again, I think, the key to the pursuit of this line of business for us is we believe it gives us diversification and exposure to the consumer segment, and we said this in the comments, in a way that's consistent with our commitment to excellent credit quality. .

And our next question will come from the line of Jennifer Demba with SunTrust. .

A question for you, you said that demand has slowed a bit for the RESG loans, you said it wasn't geographic -- I think, you said it wasn't geographic in any way.

Has it -- has there been any difference in demand in product types that you've noticed?.

Nothing appreciable. No. .

Okay.

And one more question on the marine and RV loans, what is the average size of those loans?.

It's roughly around $100,000. .

Okay.

And how comfortable is the percentage of the total loan portfolio? Are you comfortable growing that over time?.

Jennifer, I think it will grow, we haven't set any sort of growth limits on the portfolio.

While these loans tend to be longer-term fixed rate loans, the reality is they are probably somewhere between 3 and 4 years in average life because of the fact that being high income, high net worth borrowers predominantly, these customers tend to trade up on a fairly frequent basis, and they tend to have capability to pay off these loans fairly quickly.

So it's really sort of 3- to 4-year average life project -- product.

So what I would tell is you probably see us continuing to have pretty good percentage year-over-year growth in that product category for the next couple of years and as that portfolio gets to more of a mature state, 2 or 3 years out, where we're having a lot of payoffs in that portfolio and a lot of trade ups in that portfolio, the growth probably gets into more of a long-term sustainable growth rate, but I think we've got a pretty runway for a nice percentage growth of that portfolio over the next couple of years.

.

And our next question will come from the line of Stephen Scouten with Sandler O'Neill. .

You guys have always taken a pretty long-term view on the stock, which I appreciate. I'm wondering if, given where the stock is, it's currently trading, and the fact that you mentioned you are no longer subject to DFAST, and the capital raises, I think, we were all kind of expecting would need to occur to meet those hurdles.

Do you think about a share buyback of a material proportion at this point? Or is the dilution there, day 1 dilution something you'd be hesitant to undertake? And if so, I guess is there a -- what's the delimiting factor in terms of a capital ratio for you guys moving forward?.

Stephen, good question. I would tell you all options are on the table for consideration at any time in how we manage our capital position.

Obviously, we expect, as we've said in the past, to achieve good growth in the future, we'll have times when that growth will be higher than others and if history is a guide to times when we have the highest growth or times when other competitors pull back from various spaces and provide us an opportunity to really do thoughtful intelligent things in a meaningful way and volume.

So we expect to continue to need some of the excess capital we've got over what's well capitalized or adequately capitalized plus capital conservation buffers today.

If you look at the management comments, Tim, what page was that, that we put that on, 30?.

34, figure 32. .

Page 34, the management's comments, figure 32. We had a very nice increase in our capital ratios in the quarter just ended and part of that was because of the reduction in capital allocated for unfunded loans as that unfunded balance went down.

Part of that was the new DFAST, RV, new HVCRE clarifications that came as part of the Senate Bill 2155 that was signed into law. And part of that was just our organic growth and retained earnings and capital retention after the dividend. So we have a generous capital position.

We would expect to maintain a good cushion over what would be the fully phased-in Basel III standards for a well-capitalized or adequately capitalized plus the capital conservation buffers.

Within the limits of those constraints, we've got options to consider all alternatives and one alternative that could possibly be considered would be issuing some other form of capital to replace some common stock to some extent that would be accretive to our EPS number so we're looking at that, we'll continue to look at that and make what we think are the appropriate decisions, balancing all factors.

.

Okay. Makes sense.

But your view would be you wouldn't necessarily bleed that 11.9% CET1 down to like 10.5%, it would be more of a replacement in the capital stack with some other form of capital?.

Again, there are a lot of variables and a lot of options there. One option would be to replace some common equity if we did a stock buyback with either sub-debt or preferred or some other form of capital that the swapping of which would be net accretive to EPS.

But we're not saying we're going to do that, we're just saying we're evaluating all options. .

Yes. Makes sense. And then maybe thinking back about the core spread.

One question I had on just the moving parts there, was there any sort of material impact within the non-purchase loan yields from deferred origination fees, kind of quarter-over-quarter? I know that was a somewhat of a negative impact last quarter in expenses and other things, and so I was wondering if, even though it moved up 14 basis points, with some of that quarter-over-quarter impacted by the deferred origination charges, and just with that seeming dislocation between the move in LIBOR and Fed funds, I guess how can you help me think about that? Or am I just making something that's not really there?.

Stephen, I don't think there was a material impact one way or the other from prepayments in the nonpurchased loan portfolio or the impact of cost deferrals or anything, I mean, that was all -- both quarters had a fairly normalized rate, I believe.

We do have impacts every day to some degree, a few thousand dollars or a few hundred thousand dollars, some days plus or minus from loans that prepay sooner, that paid earlier in the life of the loan where there were deferred fees or deferred cost or a net of those credit or debit that impacts earnings.

But both Q1 and Q2 were pretty typical in that regard. .

Okay. Great. And then last question for me, I apologize if I missed it in here somewhere.

But the amount of offices in spin-up mode, did you give a number for that? Or where we are relative to where you were maybe last quarter?.

Tyler, do you have that?.

I do. Total bank-wide currently, Stephen, are 44 offices. So about 18% of the total offices that are in spin up today, and the specials we're offering in those offices range from 13 to 15 months, from a 2.01% APY to a 2.30% APY. .

Okay. And what was that number, I'm trying to look it up, was it like 37 offices last quarter that were in spin-up mode or... .

Yes. It's pretty close to that. We've been in that range for several quarters now. .

And our next question will come from the line of Arren Cyganovich with Citi. .

I was just thinking in terms of potential slowdown in loan growth, with your ability to dial back your operating expenses.

Are there any kind of levers around that to help maybe offset some of that from a operating leverage perspective?.

I would say that is fairly minimal. We, of course, operate with probably an efficiency ratio already that's probably in the top 1% or 2% of the industry. So we don't have a lot of fat. And we pay our lenders to make loans, and we pay our lenders to not make loans.

In buy back, our lenders are instructed to make every good quality, good yielding loan they can, consistent with safe, sound and prudent banking standards. We articulate 8 rules for them, and we say that if you can't achieve all 8 of those objectives on every loan, don't make a loan.

So if we've got a lender who is operating in an environment, and he looks at 100 loans over the course of this year and makes 0 because competitive dynamics have made it impossible for him to make loans that meet all of our standards, then we're going to reward that guy for his discipline and his hard work in creating 100 opportunities, and not punish him because he didn't produce any loans, if he worked hard and went out and pursued our standards.

And you don't want to cut that person just because he didn't make any loans this year, if he is doing his job well and following our standards because next year or the year after he may be the guy that makes 30 of those loans.

So we do take a very long-term view of this, and our lenders understand when volume is a tertiary consideration and quality is #1 and profitability is #2, you have to be consistent in the way you approach your lenders on that because they've got to understand that not hitting a certain volume metric is okay, as long as you're doing the work, you're working hard, you're producing opportunities, and you're not hitting the volume metric because you're adhering to our standards for quality and profitability.

So that's just inherent in our culture. So there's not a lot of variable costs we're going to take out if our loan growth is 75% of last year's number or 125% of last year's number, it's going to be pretty much the same cost structure. .

Okay. And then in terms of the competition for the RESG, is it -- are you seeing it more on pricing or structure, I'm just trying to get an idea, there's been some concern from investors that the return hurdles may start to go down a bit because you're giving back some of the benefit from tax reform.

I'm just kind of curious if you're seeing it more on the pricing side?.

Both. As we said in our management comments, we've seen increased competition in both structure and pricing. .

Okay.

And then just a quick modeling question, do you happen to have the purchase accounting accretion for the quarter? How much that was?.

Yes. It was roughly the same as it was in the first quarter, which was a $12.6 million recognized during the quarter. .

And our next question will come from the line of Michael Rose with Raymond James. .

I wanted to get back to this quarter's provision in reserve, and how we should think about it as we move forward.

I assume obviously you're building in higher level reserves for the RV and marine business, and that was a good portion of this quarter's growth, but as we get through 2019, and how should we think about the trends in the reserve ratio, assuming that you still have pretty solid growth in that business and maybe RESG is a little bit slower? And then if you can give us your initial thoughts on what the impacts of CECL will be for you?.

Well, it's too soon to comment on CECL, I think, Greg, would you agree with that? I mean, we're doing work on that, but that's a 1/1/2020. .

Yes. I think, from CECL standpoint, I'll let George talk to the first question, on CECL. I would say that wherever that ends up, whether it's the same allowance, less allowance, more allowance, we're still working through that analysis and trying to determine where that's going to be.

We're probably still 3 or 4 quarters before we really begin to have some [ modeling ] results on that, Michael. But regardless of where that ends up, that's kind of a onetime, and in our minds a onetime kind of reset of the allowance, as a percent of that portfolio.

I think, our thoughts are, at this point in time, as you think about that moving forward, though, from that point once you implement CECL, I don't anticipate there being a material change to our quarterly provisions, I don't think CECL is going to require us all of a sudden to start booking significantly more, significantly less provision, assuming all else is the same.

So I think it's a kind of a onetime implementation adjustment, assuming it is an adjustment, and then I think the ongoing provision is again all things being equal, not going to be materially different than what we've been running over the recent quarters. .

Yes. And, Michael, what I would tell you on the provision expense, I mean, if you look back on page 13 of the material that was part of our press release where we've got that 8 quarters of data, our provision expense in the quarter just ended was $9.6 million, that is not the highest in the last 8 quarters, the highest was Q4 '16, $9.855 million.

We had another $9 million quarter in Q4 '17 and those numbers have kind of varied from just under $5 million to just over $9 million and I think quarter-to-quarter, those are probably indicative of ranges we would expect to see.

We're not expecting a fundamental change in economic conditions or portfolio conditions that would dictate a wide variation in that provision expense. .

And to George's earlier point, we -- the RESG portfolio, we've got the history there.

We understand, we've been doing that for years and years, the allowance we carried for those loans, those very low leverage loans, it's probably at the low end of the loans we carry will be for consumers for all business and other sections of portfolio just because we have the knowledge, the history, the performance there.

The marine and RV is a little newer sector for us, and we feel really good about the credit quality of that sector as well, but we do carry a little higher allowance as a percent of loans in that sector. .

Okay. That's helpful. And then maybe back to the expense question that was just asked. I know you guys have spent some money to build out systems and things of that sort to become a much bigger bank.

Where do we stand with that? And could we see some expense leverage in the back half of the year? Or is the money that you've spent over the past few quarters, will that be spent in other areas, meaning that we shouldn't expect to see kind a net reduction as we move forward?.

Yes. Michael, I think the comment we gave on that in one of our previous calls recently was as that we would expect to see a continued build of human infrastructure to some degree.

Most of that's in place in our company, but we will continue to add some people doing some things related to this buildout of IS, IT, cybersecurity, all the elements of risk, credit review, internal audit and so forth.

We've got a few more people to add in, in those disciplines, but -- and other similar disciplines, but we do hope that those additional cost increases will be offset by reductions in our consulting expenses.

We've been spending a lot of money on consulting work, building out a lot of this infrastructure, and we expect and hope that that's going to moderate as we get more of our own people in place doing those functions, and get them more effective and have less of that infrastructure build. .

Okay. That's helpful. And then maybe just one last one from me, just on the growth in the securities portfolio. If I remember correctly, I thought you guys were just about done with that, but then the commentary implies that there could be more to do.

Any sense for magnitude?.

No. Not really. Part of that will be based on opportunity, I'm sure our regulators would always like us to hold more liquidity. Our shareholders would like us to hold less liquidity because the margins on that liquidity are not as good as other elements.

So it's a balancing act of achieving management and shareholder desires in regard to returns and meeting our internal requirements and regulatory expectations regarding liquidity.

And as our balance sheet grows, we will have to keep additional liquidity on balance sheet and we know that, and expect that and the magnitude of that is not as clear to us, but we know the direction is toward having more liquidity on the balance sheet. .

Okay. Maybe just one last quick one from me, I'm sorry. Back to Stephen's question about kind of shareholder value and unlocking it.

Given where you guys are trading, would you ever consider a sale? I mean, does that ever come into play? You're at a point where evaluation, where somebody could actually make the math work, particularly in light of your growth.

Just as we think about unlocking shareholder value, is that in the cards at all?.

Well, that's not -- it's not a -- certainly not a focus that we have, Michael, but at the same time, we're a publicly-traded company, and we've got to do the right thing for shareholders and we understand that.

So if offers came our way and somebody offered us a substantial premium, our board would have no choice but to fairly and thoroughly consider the option. That's not a direction we're pursuing, not a direction that I think we'd go.

But again, our board has got an obligation to consider all options, just as we would consider all options regarding the components of our capital structure. .

And our next question will come from the line of Brock Vandervliet with UBS. .

George, if we could just circle back to the growth issue. Is -- I just -- I find it -- and maybe you could talk about the unfunded commitments because it seems like that -- it's such a massive number that should kind of carry the day, carry the rest of the year.

I'm curious what's happening there? Is that what you're talking about in terms of transactions just don't pencil, given higher expenses, as people now look at them? Or what's happening there in terms of that pull-through?.

Brock, I'm not sure I understand your question, but let me try to intelligently respond to it even if I don't understand it.

The unfunded balance of our construction loans, which is about 97% probably of our unfunded, 93% of that is RESG and there is probably another 3% or 4% of community bank that is construction loan, the other minute part is other types of lines credit and so forth.

But the vast majority of that unfunded is construction loans and in the typical loans, the vast majority of the loans, we're the last dollars in the transaction and the first dollars out.

So if we have a $50 million loan on a $100 million transaction, all of the $100 million of equity or mezz debt or whatever the other capital components will typically be in the loan before we fund, and we'll fund the last dollars in.

So if you've got a construction loan on a project that's simple and small and easy to build, if that construction loan may fund up in 9 months or 12 months or 15 months.

If you've got a construction loan on a really large complex mixed-use project and we're at 30% or 40% of the cost of that, we may not fund till the second half of year 2 or year 3 or even beyond.

So the reality is that, I think, your point is why, if you've got $12 million in unfunded commitments, $12 billion in unfunded commitments, that's going to take care of your growth for the year, but the reality is that $12 billion will not all fund, probably somewhere 85% to 90 something percent of it is going to fund.

And what does fund is going to fund some this quarter, some next quarter, some each quarter of next year, some each quarter of 2020 and they will be little tails that will run out even beyond that.

So it's a -- it is a long-term realization of those balances being on our books and at the same time, we've got a constant rate of payoffs and constant maybe is the wrong word, an ongoing rate of payoffs, there was near $1 billion in Q1, I believe, and $1.4 billion in Q2 and that's just from the RESG portfolio, that doesn't count the other portfolio.

So there's $1 billion to $2 billion of loans getting paid down every quarter as you fund up your unfunded balances on those construction loans that are already booked, and as you make new loans to create new volumes.

So we've got a very detailed process for managing and projecting that and that the RESG, it's a loan-level prediction of how much funds every month, when a project should stabilize, when it should pay off, what the sell-out rate will be on it and so forth? So we are showing on the loan level at RESG, on closed loans and loans expected to be closed, what funding on those is.

What the pay down on those are expected to be. And then based on pools of loans from elsewhere in the company, we're looking at that on a portfolio and pool level of loans. So we have pretty good intel that tells us how much liquidity we're going to need in any given month for the next 36 months to fund that.

Those projections are not perfect and obviously, they're much better over the next 3 months, than they are in months, 12 through 15, and 12 to 15 are much better than months 24 through 27, but we're constantly re-projecting that number, looking at a 36-month forward funding forecast.

So hopefully, that answers your question about how that unfunded works. .

It does.

Is there anything you can share with us in terms of the granularity of the RESG portfolio in terms of, like, 10 largest commitments or X percentage of the entire portfolio or something like that?.

I don't have that data at hand, I think our average loan size last year was about, Tim, $77 million, is that right? $80 million?.

I think the average origination was around that. Average in the portfolio is closer to $60 million. .

Okay. So that -- there's approximately in the RESG portfolio, there are approximately 350-ish credit loans plus or minus in that portfolio. And [indiscernible] $77 million or $60 million is an average origination in that portfolio, average that's funded and unfunded [indiscernible]. .

And our next question will come from Matt Olney with Stephens. .

George, I want to stick with Brock's questions on the RESG portfolio and the dynamics behind it. I think it would be helpful to put some numbers behind it, if we can understand some of the puts and takes.

I believe that you disclosed that the RESG pay downs were $1.4 billion in 2Q, but could you also disclose what the amount of originations were from RESG in 2Q? And then what the level of RESG advances were in 2Q as well?.

Tim, you want to take that?.

Yes, Matt, the advances were roughly $34 million more than the repayments, so... .

That was the net growth, unfunded. .

So that was the net growth. And then they had roughly $1 billion, over $1.2 billion of originations there in the quarter. .

And Tim or George, that $1.2 billion of originations, you talked about how you're sponsored to be more selective on some of the deal flow, how does that $1.2 billion compare to previous quarters?.

Q1 was, Tim, what was Q1?.

Q1 was right $1 billion and the average last year was over $2 billion, per quarter. .

Now -- so if you look at that, you say okay, we're running at about half the origination volume at RESG we were a year ago, and on an average basis that would be true. The reality is, a higher percentage of our originations occur typically in the fourth quarter of each year and that's certainly been true for the last 2 years.

So we would expect some boost in that origination volume as we go through year-end. Part of that is a lot of the competitors that we deal with get annual allocations. So they've got $700 million of apartment loans or $1 billion of office building loans they're going to make and their originators are paid bonus based on achieving their goals.

So a lot of our competitors come out of the shoot very aggressive, early in the year they get their quotas, they get their allocations bill and they tend to go by the wayside.

So our volumes in Q3 and Q4 have traditionally benefited from the fact that a lot of our competitors get out of their space after they fill their allocations in, and our guys work 7 days a week, 365 days a year, and we run through the tape at the end of the year, so we tend to get an extra boost in volume in Q4 and hopefully, this year will be that way again as it has been in prior years.

.

That's helpful, George.

And then as far as the $1.4 billion of RESG pay down, I know we've been talking about pay downs for well over a year, it seems like, but how does that number in 2Q compare to the more recent quarters?.

Well, as I said, it was a $1.4 billion in Q4 of last year, it was $1 billion roughly in Q1 of this year, and I think $1 billion was sort of the high-end before Q4 of last year. So we have hit a higher number and that -- that just reflects the fact that we've got a lot of growth in that portfolio in recent years.

And a lot of those projects that we booked 2 or 3 years ago does sort of put some color on Ken Zerbe's question and Brock's question. A lot of those projects we booked 2 or 3 years ago are stabilizing now and are either refinancing their permanent market or their condos or other sorts of development loans are selling out.

And that's why we said, we expect in Q3 and Q4 our run rate of pay downs and payoffs to continue to be elevated as we've got a lot of projects that are going to CO over the next 2 quarters and sell out and we know that exist there.

So we'll have to overcome that loss of volume with new originations and funding on the existing loans and that's why we've given the guidance we've given. Some of those projects are selling faster and are tending to move to the permanent financing faster than in the past. And that is tending to accelerate the realization about those pay downs. .

And our next question will come from the line of Brian Martin with FIG Partners. .

Just 1 question back for that payoff question you just talked about, but the payoffs, I guess, if you look at the next 2 quarters where you've got maybe more visibility on those payoffs, I mean -- is -- and you've talked about this in the past, I just don't recall, but, I guess, if you look at next year on how you guys are thinking about payoffs, is there any impact from rates going up that slows down the payoffs? Or is it just based on when you booked these projects and kind of how you see them unfolding? Just trying to get a gauge for how we should think about payoffs in '19 with a little bit more color behind kind of what's going on there, so?.

Yes. Brian, our thought is and our history seems to have borne this out as being about [indiscernible], but our thinking is that as long as rates are expected to go up, sponsors on income-producing properties are going to try to refinance as soon as they can.

And the reason being, if you're paying us 6% on a construction loan and you can get permanent financing that's twice as much money at 5% today that's fixed long term, you want to get out of our 6% loan and get to the 5% and you want to do it as quickly as possible because you think rates are going to go up next quarter and the next quarter and next year, and the 5% permanent loan may become a 5.25% and 5.5%.

So the fact that the expectation is that rates are going up is encouraging sponsors to refinance long-term fixed rate faster than if they thought rates were going to fall.

And when we get to what we think is the peak of cycle, we would expect loans to stay on the books longer because then a sponsor may be paying us 7% on a construction loan, may be able to go again at twice the dollar amount at 6% on a permanent loan, but he's thinking, well, if I wait 3 more months, permanent loans rates are going down, I can get refinanced at 5.75% or if I wait 9 more months permanent rates may be 5.5%.

So when we reach a point where the expectation is that rates are going down, the assets will tend to stay on our book longer because the sponsors will want to maximize their exit into their permanent financing.

Now if you're talking about condo sales or lot sales or single-family home sales, those are not going to be affected much one way or the other by rate expectations. .

I got you. That's helpful. And most of my other stuff was answered. Just maybe one other thing, just going back to expenses for just one question.

And that was when you look at -- last quarter, there was the deferred cost you talked about being an impact, obviously, and then the infrastructure build which appears to be slowing or at least kind of being remixed a little bit.

But the -- is -- would you consider this quarter's expense run rate at pretty clean level? I understand you're still investing in the company and getting some benefit on the consulting side, but just where it sits today was more representative of kind of a cleaner quarter from an expense standpoint, and how we think about it?.

I would say, yes. Absolutely. .

Brian, I would say that professional fees is still a little elevated in the quarter.

We would certainly hope that we can continue to pull those back down, but they will likely be replaced with additional headcount and resources as we kind of swap the consulting and professional fees for the personnel, but I do think that the overall level of the overhead is a pretty good starting point to look forward. Yes. .

Hey, Brian, I like the quote on your firm's daily note today. .

[Operator Instructions] And our next question will come from the line of Blair Brantley with Brean Capital. .

Just a couple of quick questions.

Everything else has been answered, going back to pay downs, on the purchase book, is that -- are those pay downs as expected? Are you expecting a $25 million or $50 million drop each quarter? Kind of where it has been trending or how should we think about that -- and -- in the view of net loan growth?.

Yes, Blair, this is Tim. Hard to predict obviously on those, but obviously, our portfolio of purchase loans is going down.

So the dollar amount of that should go down, whether we have 10% per quarter or 12% per quarter pay down, I don't know that the percentage change -- decline, changes that much but obviously the dollar amount, that change, should decline as the portfolio declines. .

Okay. Great.

And then just on the fee income side, was there anything that kind of was non-recurring this quarter?.

No. Not really. No. We've got -- we're always going to have recovery from purchase loans. We're always going to have OREO gains, those are all onetime in nature, but we have a lot of those that we'll have in the -- so no, I think it seems like a pretty good run rate. .

And then so even on the loan service and maintenance side of it, there's often -- that has just got to -- -- that is going to bounce around as well? A little bit?.

Well, we're getting more and more of those fees in the deals that we do today in RESG, so that has a tendency to grow over time but nothing meaningful from a onetime item in that particular line item. .

And I'm showing no further questions in the queue. So now... .

Thank you guys for joining our call today. I want to point out that next Monday, we will be changing our bank's name to Bank OZK and as of Monday, our stock will be trading under a new ticker symbol, OZK.

Our new name reflects both our rich history and our commitment to being the leader in technology and innovation as we continue to grow our company and we're looking forward to moving forward in a very positive way under the Bank OZK banner. Thank you, that concludes our call. Thank you for joining us. We look forward to talking with you next quarter. .

Ladies and gentlemen, thank you for your participation on today's conference. This does conclude our program, and we may all disconnect. Everybody, have a good day..