Ladies and gentlemen, thank you for standing by, welcome to the Bank OZK Fourth Quarter 2019 earnings conference call. At this time, all participant lines are in a listen-only mode. After the speaker's presentation, there will be a question and answer session. To ask a question during the session, you will need to press star then one on your telephone keypad. Please be advised that today's conference may be recorded. If you require any further assistance, please press star then zero to reach an operator. I'd now like to hand the conference over to your speaker today, Mr. Tim Hicks, Chief Administrative Officer. Please go ahead, sir.
Good morning. I'm Tim Hicks, Chief Administrative Officer and Executive Director of Investor Relations for Bank OZK. 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, and Greg McKinney, Chief Financial Officer. We will now open up the lines for your questions. Let me ask our Operator, Liz, to remind our listeners how to queue in for questions.
As a reminder, ladies and gentlemen, to ask a question, you will need to press star then one on your telephone keypad. To withdraw your question, press the pound key. Again, that is star then one if you'd like to ask a question at this time. Our first question comes from the line of Kenneth Zerbe with Morgan Stanley. Your line is now open.
Great. Thanks. Good morning. I guess maybe we can start off in terms of that substandard credit that was formerly watchlist. Can you just help us size the potential losses on that, and also the timing of any potential resolution around that? Thanks.
Good question, Ken. Thank you. We do not have any thoughts on what a loss might be. Obviously, the loan is still performing as an accruing loan. The fact that it's a performing accruing loan is a result of our analysis that projects our future cash flows, interest, principal repayments and so forth on the loan. While the margins are very thin, we currently project that there will be enough cash flow from the project to repay all of the principal and all of the interest on our loan. Of course, we're projecting interest into the future on that using a forward yield curve as a proxy for what interest rates will be in the future. Based on that, there is no present loss in it.
If you look at the appraised value in our bubble chart, in our management comments, it is more than 100% loan-to-value, but obviously the appraisal uses a higher discount rate than the effective rate on our loan. If you use the effective rate on our loan in a forward yield curve, there is no present loss exposure. What would cause there to be loss exposure and would cause this credit to move from substandard accruing to a non-accrual status would be a change in sales prices, projected sales velocity, interest rates, that in some combination of those, cause that forward projection of net present value to become negative instead of a positive differential over the loan amount. It's certainly, since there's no present evidence that the loan is impaired, it's premature to talk about what the loss would be.
The timing for resolution, your other question, I think this loan will be with us for quite a while. The sponsors are working the project very effectively, and while their fourth quarter sales and signing of sales contracts were a little bit below what we would have hoped for, which caused the, or contributed to the downgrade, the reality is they're still selling townhomes and selling lots. They're still starting new townhomes and working toward development of an additional small phase of lots. I think the expectation we have is this is going to be a long-term deal, and they're going to continue to work it, hopefully successfully. Hopefully their sales prices and sales velocities will be stable to improving, and that will lead the profile of this credit to improve.
If their sales velocity and sales prices decline, that will lead the profile of this credit to decline. That's about all I can say about it.
Got you. I guess in the release you talked about sort of the reason why it was downgraded was because projects were being delayed or canceled, which I guess I kind of think about as being synonymous with sort of a deterioration in sort of the cash flow outlook. Is that the right way of interpreting it, that there was a deterioration in this credit which led it to be downgraded, or was it downgraded for some other reason?
No, you're exactly right. The sponsor had several.
Contracts fall out. Some of those were for lots, some for townhomes, and they were either fell out or delayed in closing, and that resulted in several million dollars less in sales in Q4 than we had previously expected. We mentioned in our management comments that the sales volume of lots under contract that would have near-term closings was very low. I think there was one at the end of the year, and I think they've signed up one since the end of the year on the lot side. The townhome sales are better than the lot sales. That, having a few contracts fall out in Q4, lowered the receipt of cash there and pushed out the timeframe for the development sales, and the margins got thinner. The margin for error got thinner as a result of that elongation of the sell-out expectations. You're exactly right.
It's a cash flow issue.
Got it. Okay, maybe switching gears just a little bit, in terms of the RESG portfolio, I noticed that you did say that both payoffs and originations should be a little bit higher in 2020. It seems that net should still have positive growth in RESG, just kind of wanted to get a sense, is it possible that these balances could be either relatively flat or even down sort of on a point-to-point basis in 2020, given payoffs?
Hey, Ken. This is Tim. I think I'll point you to Figure eight on our management comments, which is on page nine. There, we show the trends of our originations by year and the trends of the remaining loans outstanding by year. You can see in 2016 originations, we had $8 billion of loans originated. In 2016, $2.08 billion are still outstanding. You can see in 2017, we had $9.1 billion of originations, where at the end of the year, we had $6.06 billion of those balances remaining. Those, as we've talked about before, our construction loans typically average about three years in life. Somewhere two to four years is the span, typically, but the average is around three years. If you look at our 2017 origination volume, most of those loans will come to completion this year.
As we've seen in many of our RESG loans is, once the project is complete, we get paid off pretty soon after completion. It's just a cycle of origination volume that we had from three years ago coming through. The total funded balance will move up. We may have a quarter or two throughout the year where it's like it was in Q4, which it was down in Q4, and there will be quarters where it's up. I think our good origination volume that we saw in 2019 and our expected good origination volume that we're expecting in 2020 should help alleviate some of those payoffs that are coming from our previous origination years.
Got it. Okay. All right. Thank you very much.
Our next question comes from Stephen Scouten with Piper Sandler. Your line is now open.
Hey, good morning, everyone.
Good morning, Stephen.
Appreciate, obviously, all the detail you guys give, and Figure eight is kind of where I wanted to focus as well. I'm thinking about the forward growth, and I get why pay-downs will be higher with the 2016 and 2017 originations, but it seems like those pay-downs would start to abate in the back half of 2020 as we move further through that pipeline. Is that at all possible, and do you think there's some likelihood you could see growth in RESG pick up in the back half of 2020 or 2021, or is that just too early to say?
Stephen, I would add a little additional color. I think Tim took the last question to the right figure, which is Figure 8. The majority of probably the remaining originations from 2016, a large percentage of that will pay off, a big chunk, although we wouldn't expect near all of the 2017 originations to pay off in 2020. We'll have a little bit of the 2018 originations that have paid off. One of those loans is already paid off. You will see a high level of payoffs, we expect, in 2020. Those results will be fairly variable from quarter to quarter. If you look at the first quarter of this year, I think RESG had net funded growth of about, if I'm right, $442 million. It shrunk in Q2, $228 million.
It had funded growth in Q3 of $256 million, but shrunk in Q4 roughly $157 million. We had two quarters of positive growth, two quarters of negative growth. For the year, RESG's funded balance grew about $314 million. To kind of go back on Ken's question, it is. You could paint a scenario where we would have negative RESG growth for the year. We don't think that's necessarily the likely scenario. You could also paint scenarios where we had better growth in RESG in 2020 than we did in 2019. But I think the center line of that growth is probably somewhere plus or minus, not terribly far from what we saw in 2019. Because, again, we're going to have a big wave of payoffs, and we should have better originations in 2020. We're also expecting bigger payoffs, so it probably is pretty much offsetting.
Okay. Very helpful color. Then on the NIM commentary, I guess from page 15, I was a little bit surprised to see that it sounds like even in an unchanged rate environment, that you'd see additional downside to the NIM, just maybe not the same magnitude. Can you help me with that? Are you guys giving any kind of numerical guidance around what you think the magnitude of the incremental compression could be, even in a flat rate environment?
You want to take that, Tim?
Yeah. Hey, Stephen. We gave you some comments around our expected increase in deposit cost on our commentary. You had referenced page 15, which is a good reference as well. On page 17, we talked about our cost of interest-bearing deposits. It was down 12 basis points in Q4, which followed a six basis point decline in Q3. We did indicate that we didn't think Q1 decline would be as great as the Q4 decline. We do expect it to decline in Q1. On the deposit funding side, that will help. We'll continue to work on our deposit mix and hopefully continue to improve our deposit cost as we go throughout the year.
On the loan side, obviously, if the Fed doesn't move rates in 2020 and LIBOR stays fairly stable, the impact that we would be dealing with on the loan yield side would come from competitive factors. Obviously, we're in a very competitive environment for loans right now. The competitive factors for loans and then a slight change in mix. Obviously, our RESG loan yields are higher than the average and our community banking and indirect lending loan yields are below the average. I think that's what we're dealing with on the loan yield side. Mostly from competitive factors from that perspective.
Okay. Helpful. One last clarifier for me. This substandard, obviously, you gave good color. That feels pretty well contained. I'm wondering if you could give any insights into any new credit migration, if there is any. Like where you guys list the moderate bucket, maybe in your 10-Q. Can you give us any visibility into any early-stage migrations that may or may not be occurring?
Stephen, I'm not aware of any other than our watch credit category went down by a comparable amount that our substandard category went up. That will show a positive migration, a decline in that balance. I'm not sure of any major changes between our other categories, moderate category, or any of our other categories.
Okay, perfect. Very helpful. Thanks for the time, guys.
Our next question comes from the line of Daniel Mannix with Raymond James. Your line is now open.
Yeah. Hey, guys. Good morning.
Good morning, Dan.
Wanted to dig a little deeper into loan dynamics, specifically on originations. Have you seen any change in the approval rate on your loan pipeline recently? I think it's been about 5% in the past. Just trying to get a better sense on whether or not you're passing on more loans due to competition or maybe some other factors.
Daniel, we don't track that the same way we used to in regard to flow coming in, so I can't really address that percentage number. I would tell you, we are, as Tim alluded to earlier, in a very competitive environment. That seems to be true for all types of loans, RESG loans, indirect loans, community banking loans of various types. It's an environment out there where volume is desired by a lot of lenders. We're seeing a lot of lenders get very aggressive on credit and very aggressive on rate to get that volume. As we have commented repeatedly and consistently, we're non-negotiable on our credit standards. We'll give a certain degree on rate, but not beyond a point. Our giving somewhat on rate has contributed to declining loan yields.
Our not giving at all on credit, not giving beyond a certain point on rate has contributed to our declining loan volume. We continue to believe that that discipline is definitely the right approach. We're not going to waver from that discipline, and we think we'll get rewarded for that when economic conditions reach a point where guys who are being too aggressive get punished for it. We think we'll be in a great position to shine and grow in a meaningful way at that point in time.
Got it. Thanks, George. In terms of loan demand, you guided to slightly stronger originations in 2020, but still off from peak levels from a few years ago. Can you tell us what's driving that? Is it increased loan demand in major metros, or is this a case of gaining share of a smaller pie, if you will?
I would tell you that we're seeing less origination volume in certain markets where you've got an adequate amount of supply. New York would be the poster child for that, probably. The New York market, there's a need for less new product there because there's been a lot of product built, and tax and other issues there have diminished the need for a lot of new product. We continue to originate some new volume in New York, but our total commitments in New York at the end of the quarter just ended were the lowest that they've been since the first quarter of 2018, so lowest in eight quarters. I would expect that our total commitments, funded and unfunded to New York, will continue to decline, not because we wouldn't originate good new loans there, we will, but our payoffs there will exceed originations.
In the quarter just ended, we originated loans in a lot of markets, Boston, D.C. area, and more markets such as Dallas was, I think, our second-largest volume of originations in the quarter just ended. Boston was the largest. Chicago was third. San Francisco was fourth. Atlanta was fifth. You get down and you've got Sacramento, Phoenix, Savannah, Philadelphia in the top 10. Markets that we got a lot of volume in the 2016, 2017 timeframe, Miami and New York, are farther down the list. You've got to go to the markets where the supply-demand metrics make sense and the projects that are getting done that make sense. That is, in a lot of cases, more secondary markets or at least not your traditional kind of top five markets in some cases. We're finding the volume.
My compliments to our loan team for what I think is incredible work that they did in 2019, originating $6.48 billion of loan originations in RESG, up about a billion and three quarters from the prior year, and still holding very steadfast to our discipline and pricing standards. To accomplish that, they had to burn a lot of shoe leather and make a lot of calls and really study and explore and understand a lot of markets that we got some volume out of that was very helpful to us, high quality, good yielding volume.
It's a good outcome when you look at the results that our teams achieved and the context of the competitive environment in which we're operating, and the fact that some of the major markets where we've, in recent years, gotten a lot of volume, just didn't have as much new development that created new demand.
Great color. Thanks, George. That's it for me, gentlemen.
Thank you.
Our next question comes from Catherine Mealor with KBW. Your line is now open.
Thanks. Good morning.
Hey, good morning.
I wanted to see if we could circle back to loan yields. Can you provide us just generally where current loan yields are in the community banking segment, and then in the indirect RV and marine, as we kind of think about how that mix, even if we're in a flat rate environment, how just that mix change may impact loan yields this year?
In the community banking environment, we've got so many different verticals there in community banking, so many different types of loans and specialty types of loans we do. Those yields tend to be all over the board, depending on that type of loan. I would comment in the indirect area that you guys noted, I'm sure, that our volume of growth in indirect was probably the lowest in the quarter just ended. That has probably been in a number of quarters. That reflects the fact that that market has gotten very competitive. If you follow the marine and RV manufacturers at all, you'll get the impression that a lot of those manufacturers are shipping less product because there's less product being sold at retail.
The fact that there's less consumer paper being originated on fewer marine and RV sales than there were a year ago, is resulting in less paper for lenders to the retail customers such as us and a host of competitors. Some of our competitors have gotten pretty aggressive on both credit and rate. Once again, we're having to be very disciplined on the credit to make sure we get what we want. We're having to give a little ground on rate to stay in the game. That is an area where competition has hurt our margins a bit.
Is it fair to assume that the indirect RV and marine growth will kind of remain around the level you saw this quarter? If so, is community banking momentum enough to offset that, so those two pieces are kind of offsetting?
That's a really good question, Catherine. Number one, we do expect our community banking units to be a bit of strength to us in origination volume. Those guys seem to be gaining traction, and it's been slow elevation of their production, but they seem to be continuing that trend. The indirect marine and RV space is very competitive. Now, one thing that we are doing is, in conjunction with CECL implementation, which, of course, occurred on January 1 of this year, we have built out a series of scorecards over the last couple of years for all of our different loan types. Where we used to have a dozen or fewer risk ratings for loans, we now have 72 risk ratings for loans. All that is being implemented in connection with CECL.
It lets us grade and refine our credit assessments of loans much more precisely than we have with our models and tools in the past. In addition to that, we've been building up large pools of data that we're using in those loan risk ratings and grading assessments. We're beginning to use a lot of that enhanced data that we've built over the last couple of years, and particularly the last year, into our credit analysis in our indirect lending, as well as other categories of lending. I think that part of our growth equation for indirect lending for 2020 will depend on how effective we are in utilizing our enhanced data and modeling and analytics capabilities in that area, which we had a lot of data already, and we used a lot of analytics there previously, but we've refined and enhanced all that.
I think our ability to grow that unit equal to or slightly more or slightly less than last year will depend on a combination of, one, competitive conditions and two, our ability to use these enhanced tools to be a little more surgical and precise in our pricing and approval of credit. We're doing things that we think will help us continue to keep the volume up without sacrificing quality at all and without sacrificing our yield on those loans very much at all.
Great. You mentioned CECL. If I could just ask one more on CECL. Is there any thoughts that you can give us on how you're thinking about what the potential impact could be to the provision this year? I mean, I appreciate it'll be volatile, you mentioned that in your prepared remarks, anything that we should be thinking about as we model the provision in a post-CECL world?
Well, of course, since you're reserving for life of loans and for commitments, which you've previously not provided for in the past in the form of commitments, when you have large origination quarters, you'll have a disproportionate hit to income from CECL. We'll all be saying, hooray, we originated a lot of loans in this quarter, and we'll be crying about the fact that it harmed earnings because you put up a big provision, and then in quarters where you have low originations, you'll have the opposite impact. Yes, provision expenses will be up with CECL because you'll be providing for unfunded commitments, and to the extent those grow, that will require more provision, and you'll be providing for life of loan estimated losses, not just incurred losses as existed in 2019 and before.
Great. Thank you for all the color.
Our next question comes from the line of Timur Braziler with Wells Fargo Securities. Your line is now open.
Hi. Good morning. Thank you. If we can circle back to the substandard loan, can you provide an update on the outstanding balance and what the current reserve level is?
Timur, this is Tim. On the outstanding balance, the total commitment is $57.5 million. At year-end, I think it was pretty close to that. Within each quarter, it goes up and down. At some quarter ends, it's been $50 or $52, but throughout the quarter, it will go up and down, and it's fairly close to that $57.5 million at year-end. With the migration from watch status to substandard accrual status, it does have a 5% reserve associated with that, which is greater than the 2.5% we had allocated to it under the watch rating.
Great. Thank you. Maybe just circling back to some of George's recent commentary in the indirect portfolio, can you provide an example of what some of the competition is doing in being aggressive on the structuring of these credits?
I'm always reluctant to talk about our competitors, but again, we've seen some competitors get very aggressive on price, which has forced us to adjust our pricing to a degree. We've seen some competitors get very aggressive on credit, which we've not responded to at all. Typically, in the business, your credit is really driven by your credit of your borrower, combined with how much you're willing to loan against that. I think in recent quarters, our weighted average loan has been somewhere from about 99% to 104% of dealer wholesale invoice price, which means our customers, our borrowers, the consumers, have a fair amount of down payment, either from cash or trade-in, or some combination of those in the transaction that provides us some protection and assurance of their commitment to the credit.
We've seen some of our competitors being much more aggressive in that regard and allowing a lot of back end and soft costs to be financed in the loan. There are wide differences in how you approach credit quality in that space, and we've always been on the very conservative end of the spectrum there.
Okay. Maybe switching gears, the cash position continues to grow a billion and a half here at year-end. It sounds like loan growth is going to be a little bit slower in 2020 than 2019. How should we think about the cash position? Is there any willingness to park some of that in the securities portfolio, or is the shape of the yield curve still prohibitive in that front?
Well, there's not a lot in the securities world that is very exciting right now. I apologize to our shareholders, that earlier in the year or late last year, I failed to recognize that the 10-year at 194 was a screaming buy. That wasn't immediately obvious to me. It's hard to get excited about much in the securities world, and we have had a big focus over the last year of increasing liquidity as measured by various ratios and keeping more cash on balance sheet. That cash position will vary from quarter to quarter, but I think in general, we would expect over the next four quarters to see our liquidity ratios continue to improve and our cash position be stronger at some quarter ends than others, but generally strong.
Our philosophy has been to grow capital and increase liquidity and get ourselves in a position to take advantage of opportunities caused by economic turbulence, dislocations or whatever, when and if those situations arises. I think we're continuing to pursue that philosophy.
Okay. One last one for me, just speaking of capital or the TC here, north of 15%, I guess, what's the main prohibiting factor from starting or at least announcing a buyback? Are you waiting to see the impact of CECL? Is there something else coming down the pipe that we might not be unaware of here? I guess, just what's the internal conversation as to why not at least announce a buyback here?
Well, I think we've given tightened estimates on CECL, and those are included in our management comments. Obviously, there is extra provision and extra reserve cost associated with CECL, but those are very manageable cost and probably not outside of anybody's expectations for a bank of our size with our level of unfunded commitments. Our estimated day one impact of CECL actually came down from the earlier guidance we'd given a quarter ago. That's not an issue, and there are no things that, I think to use your term or paraphrase your term, that we're aware of that would cause us to hold more capital. It simply reflects the fact that we believe in the long-term ability of our company to grow organically and at the right time, through acquisitions. We believe that that capital will be very useful and important to us in achieving those longer-term objectives.
We've, for those reasons, because we believe in our business model, elected to not pursue a share repurchase up to this point. That will be something that our board will continue to monitor, but we've not pursued it up to this point.
Understood. Thank you.
Our next question comes from the line of Arren Cyganovich with Citi. Your line is now open.
Thanks. With the top line growth somewhat challenged, it sounds like in 2020, you still have operating expenses guided to the high single digits. Are there any levers that you can pull from your expense side to better match the top line growth that you have for 2020, and maybe even forward as well?
Arren, there are a lot of levers that we could pull. Our focus is on really improving our company and preparing for the future. We could do what a lot of banks have done and will do in these situations in the past, and that's lay off a lot of people and cut costs and just hunker down. We are very forward-thinking, and we really believe intensely in the future prospect and power of our business model and the various business units we've built. We are continuing to invest significant sums.
We're trying to do it as prudently as we can, but we're continuing to invest significant sums to build our human infrastructure, our technology, our risk systems, and all of the other systems controls and things that you need to grow and be a larger bank than we are today and to really deploy and optimize our experience for our customers as well as our results for our shareholders. Now, we could be very short-term oriented and say, wow, I'm hyper-focused on driving every $0.01 of EPS I can in 2020, and I'm going to cut a bunch of costs and I'm going to be very stingy in paying our people, even the high performers, and risk losing some of those people and whatever.
You can take the approach that we're taking, and we believe that when we get past this wave of payoffs and get a lot of this infrastructure build really finalized in 2020, that we'll be in a very strong position to grow and advance our company in 2021 and 2022 and 2023. We can spend some money investing now so that we've got the people and the processes and the customer relationship and all the things that we need to do to grow in those years and do it in a very meaningful and favorable way for shareholders. We feel like we're very much in a short-term versus long-term decision mode here. I certainly want our 2020 results to be good.
Frankly, I'm much more concerned about building the infrastructure and improving our company so that in 2021 and 2022 and 2023, we can do great things. That's the focus. In 2019, I visited every office in our company and spent 45 minutes to three and a half hours with every team in our company. I asked two questions repeatedly in every meeting, and those questions were, what can we do to improve the experience for our customers every day, every way? What can we do to improve our efficiency and your work environment and make you more efficient, more productive, and as an employee of our company, have a more enjoyable work experience? If our staff is feeling good about our company and highly motivated, that comes across to our customers.
If that comes across to our customers, we have more success growing our business and expanding and building the kind of profitable relationships with customers that we want to have. We got literally hundreds and hundreds, I think over 1,100 recommendations for improvement for our staff that we have elected to adopt. We've already implemented somewhere close to 700 or 800 of those improvements over the course of this year. We spent some money doing that, and we'll continue to spend some money doing that in the next year. We think in the world of banking, where there are going to be fewer banks every year, that the banks that are going to be successful five years, 10 years, and 15 years from now, and there are not going to be nearly as many of them.
We think that those banks are going to be banks that build exceptional experiences for their customers, and we're working really hard on doing that. We're spending money on that and a lot of other infrastructure development. I could pull back on that and improve EPS a penny or two a quarter. I think that would be silly and foolish because I think the long-term potential of what we can achieve if we really take advantage of this slower growth period in which we find ourselves right now to really fundamentally enhance and build and improve the capabilities of our company in a significant way. I think we get paid back for that massively in future years. I'll ask our shareholders to be a little bit patient and think long term as opposed to short term.
I think if we do that, we get a good reward out there for that patience.
Thank you. Very helpful.
Our next question comes from Matt Olney with Stephens. Your line is now open.
Hey, thanks. Good morning, guys.
Hey, good morning, Matt.
George, I wanted to stick with this discussion on expenses. I appreciate the long-term focus for shareholders and not looking for shortcuts with layoffs. It also sounds like you're working on ways to become more efficient in the future. I think what we previously thought that the larger infrastructural build-out was going to wind down and slow the first part of 2020. Now with the new guidance, it seems like that infrastructure build-out will continue throughout the year. I guess the question is that we're wondering is, what's changed over the last few months?
Well, Matt, I tell you, we talked about early in last year that I was really hopeful that we could get a lot of these consulting costs that we were paying out of the company and that we were building a lot of staff members in-house to do a lot of that work, and that would allow us to get rid of the consultants. With this loan scorecard risk-grading project, CECL, a lot of the data-building initiatives that we have pursued to really accumulate, manage, and get a lot more control over data and some further enhancements that we've made to credit risk and analytics and modeling capabilities and internal audit and so forth. We've continued to spend a little more money, probably every quarter than I expected to spend to get to what was a fully developed end state of a lot of those initiatives.
We've been slower to get rid of the consulting cost than I had hoped we would be. We're taking another hard run at that in 2020. I'm cautiously optimistic that we will be much more successful in reducing consulting cost in 2020 than we were in 2019. That could help us tame some of that expense growth. We also, because I visited every office in the company last year and didn't do that solo, did that with a pretty sizable entourage of community banking people that went with me to really get a much better grassroots view of what's going on in every market in the company and so forth. We're trying to really mitigate travel expenses this year.
Not only did I go to those markets, after I went and identified specific issues in the markets and so forth, we had a lot of other people go back to work on improving, enhancing, whether it was technology, process, procedures, training, whatever, work on various things that we identified we could improve. We hope to get some of those costs down, but at the same time, we also realize we've got to continue to grow our pool of talented people, and that includes people who can originate and produce new business. We are spending a little more money than I expected to spend or hoped to spend, but sometimes things cost more to get to the end state than you originally expected. We've continued to experience a little bit of that.
We are trying to cut out unproductive and inefficient expenditures and get more efficient in, for example, doing things with our internal team, which we can do less expensively than we can with the consultants. We're working on it.
Okay. That's great commentary. Thanks for that. Switching gears, on the deposit side, we've seen a pretty big shift of deposit mix over the last two quarters. I think CDs now represent around 40% of overall deposits, and it's trending higher. Should we expect an additional shift from here? If so, what's driving that mix shift at deposits?
Yeah, I think you're seeing a couple of things there. Number 1 is, when rates were very low, a lot of people parked money in money market accounts and savings accounts because there was not much yield difference between money market and savings and CDs. You've seen a steady migration that continues even now. As rates went up and CD rates got higher, people began to be a little more judicious about how they managed that money. We've seen a lot of that trend, and we're seeing that trend continue. Secondly, we had some wholesale sources of deposits that were in money market accounts or other non-CD accounts. We are making a concerted effort to improve the quality of our deposit base, and this is part and part of our tour of all the branches and so forth.
In a lot of our rural markets, one of the keys to attracting new deposits and new relationships and building core customers is driven from the CD side of the business. We're replacing, in a lot of cases, wholesale non-CD deposits that were chunkier and probably more volatile with local market CD deposits that are much smaller, more granular, and much less volatile and have the additional benefit of giving us cross-sell opportunities for core account relationships and so forth. This is a conscious decision. While you would normally look at it and say the shift in mix from non-CD to CD is an adverse shift, we actually view that as a positive shift when you get down and you really drill into the part of it that is due to wholesale non-CD funding sources being reduced in deference or preference for retail CD funding sources.
There's a bigger strategy at work here. We think that strategy will be very beneficial to us longer term.
Thank you.
Thank you.
Our next question comes from Brock Vandervliet with UBS. Your line is now open.
Oh, great. Thank you. The problem, the downgraded credit, this is the Tahoe credit that we've talked about in the past, yes?
Yes, near Tahoe. It's near Truckee.
Got it. At this point, are all the proceeds from lot and home sales, are those going directly to repay the loan or pay the interest on the loan? Is it only a portion of that?
Brock, there's a revolver in the facility for home construction and a revolver in the facility for lot development. The proceeds basically come in, pay the loan down, and then they're building new townhomes, developing new lots from that. These are principally revolving facilities that the money does come in and pay it down, but it gets redeployed to build the next phase of the project or the next units in the project.
Has there been discussion of just locking down the principles in terms of, "Hey guys, let's not pursue the next phase. Let's wrap it up and take our lumps and move on"?
I think the much more prudent approach is these guys have decent to good sales velocity and pricing momentum. I think the much more prudent course, Brock, is as long as they're having good success doing it, let them continue to develop this thing out. This is a very nice project, and it's a very viable project. The simple reality is this project started right before the Great Recession. It got severely revalued as a result of the Great Recession and never has gotten a much higher valuation. Somewhat higher, but not fully recovered what were the original expectations on it. As a result, you just got too much debt. You have an over-leveraged project, as is clearly evident from the loan-to-value on it. It's an excellent real estate asset that is being well received by consumers.
Townhomes are selling, lots are selling, homes are being built on many of the lots that are selling. It's a very viable project. It just has too much debt in it. We think the way to maximize our credit and hopefully avoid any loss in the credit is to continue to let these guys execute the business plan and get this thing worked through. Now, there will reach a point in a couple of years where there will be no more lots to develop. There will be no more or fewer townhomes to build, and the loan will begin to amortize as a result of that. The revolvers will go down because you're not replacing existing product with new product. It's a working project.
How far away are we from that point where it's just paying down, would you say?
A couple of years into the future, and I don't know the exact date, Brock, but it's going to be here for a while.
Got it. Okay. Thanks, George.
Our next question comes from Jennifer Demba with SunTrust. Your line is now open.
Thank you. Good morning.
Good morning.
Just question about deposits. You brought in a Chief Deposit Officer about a year ago. You talked a little bit about, a few minutes ago, about CDs and getting some more retail CDs versus wholesale. What other strategies have been implemented, or do you plan to implement in terms of gathering or improving the funding mix?
An excellent question. We're approaching that from all sorts of different directions and I think making some progress. We're clearly looking at our deposit product set, and we are in the process of redesigning all of that deposit product set. We will roll those products out probably in the second quarter of this year. There's a possibility we may roll those out, some of them, late first quarter, but I think more likely it's a second quarter 2020 rollout. At that point, most of the existing deposit products will become back book products, and we'll be selling a totally new product set.
We think that is going to include a number of features that will be very much desired and appreciated by our customers that will have us in a situation where we're selling our products based on the quality and convenience and the philosophy of those products as opposed to solely on price. We think it helps us both in customer acquisition and cost of funds. We are also working to make sure that our commercial products are much more desirable and we're a much more effective competitor for that. We will be rolling out, tentatively scheduled for April, a major revamp of our commercial products and how our customers interact with I'm sorry, August. I said April. August. The other A month. That is a very technology-based evolution of our commercial products. We think that will, again, be very well received.
We're working on an evolution of our technology for our customers to interact with us on wire transfers that we will roll out in early 2021. We are working on, and have implemented really, a total redesign of our retail banking staff as a result of having visited every office in the company. Our Chief Banking Officer and Chief Retail Banking Officer have redesigned the job descriptions and job titles of every employee in the company. We made 15 acquisitions. We slotted people from those acquisitions into job descriptions that fit in our former legacy Bank OZK world but didn't necessarily match up with the cultures and staffing of some of the banks we acquired.
As we visited every office, Cindy Wolfe and Carmen McClennon, Cindy's our Chief Banking Officer, and Carmen's our head of retail banking, Chief Retail Banking Officer, and Alan Jessup, who is our head of community banking on the lending side and all of our specialty lending verticals and community banking, realized that we really had a lot of people somewhat miscast. We're not maximizing the potential.
After going through every branch and visiting every branch and really understanding at a very detailed level what's going on in the branches, the four of us and several other people went through every single employee in the company and talked about those employees with their supervisors, managers, and their managers' managers and supervisors, and reassigned job descriptions and titles for every person on the retail side of the company that we think will allow us to maximize the individual potential to contribute to our success and our customers in the company. We also in the process of revamping our call center and digital services. We've revamped, over the last year, marketing, and we've revamped what used to be training, which is now organizational learning and development, all with the goal of helping us be a much more potent force in retail banking and deposit gathering.
We're taking a very broad and holistic approach to this, as well as digging down into the details at every point of attack. I'm very excited about where I believe our company is going to go and evolve to as being a highly competitive retail bank. We've also taken all of our online apps. We went through a very significant conversion, which required us to write a lot of code to do it, and got all of those consolidated into a single app that allows us the functionality and flexibility to add various features along the way without having to burden our customers with having 5 apps on their mobile device or 7 apps on their mobile device. It all integrates through one single app, and you can move very fluidly and very quickly from one function to the other.
This was a huge undertaking and puts us in a good foundational position to really begin to roll out a much more effective and desirable mobile banking platform going forward. We've been working on a lot of things, and we spent a lot of money on these things, but I'm convinced there's a great payback that will ultimately come from this work.
Thanks, George.
Thank you.
Our next question comes from Brian Martin with Janney Montgomery Scott. Your line is now open.
Hey, good afternoon.
Hi, good afternoon, Brian.
Hey, just one question, George. Just I appreciate all the color on the infrastructure build. Just kind of thinking about the efficiency and how trends play out, it sounds as though the expense growth rate could moderate some in 2021. I guess the way to think about it is the efficiency should trend a bit higher as you continue to execute on what you've outlined and then maybe moderate some back in 2021. Is that kind of fair how to think about it? Or how much impact the efficiency could have this year with the change in the build-out?
Brian, I think given the fact that we've got very conservative growth expectations for 2020, given the payoffs that we've already talked about and the competitive environment, I think you're probably thinking about that correctly, that the trend in the efficiency ratio in 2020 is probably up, not down. If we can begin to moderate that rate of expense growth in 2021, we can get more in line with our historical growth rates in 2021 and 2022, then I think we can begin to see that efficiency ratio get better. You're exactly right.
Okay. Just one back to the loan yields. I think you talked about the community banking and the indirect, but just on the RESG, are you seeing yields there on new production kind of stabilize, or are you still feeling pressure to kind of give a little bit on the rate there for the right credit that you're looking at?
Well, I think you always, in any environment, you find yourself that for a particular credit and particular relationship, you give a little bit on pricing. One of the challenges that we've had in RESG is in 2016 and 2017, we were in a much less competitive environment for construction and development loan origination. We got wider margins. We talked about that in that period of time, that a lot of banks had pulled back from the space and we were getting wider margins. As the banks came back into the space, those margins got back toward more normal margins.
With the Trump tax cuts, we tried to hold the margins, but a lot of banks just began, and debt funds and others began to sort of bleed those margins lower, taking advantage of the lower tax rates to get more competitive on the loan side. I don't think our margins versus LIBOR or prime rate, I don't think we've seen much degradation in our margins over the last year versus that. Clearly, the margins that we were working with in 2019 were less than the margins we were achieving in 2016 and 2017. As those older loans have rolled off and the newer loans have rolled on, we've probably been replacing those at a lower margin. I don't think that's particularly changed over the course of 2019.
I think it's more of a 2019 versus 2016 and 2017 phenomenon than it is a Q1 of 2019 versus Q4 of 2019.
Okay, that's helpful. Just last thing for me is just going back to that figure eight, when you talked about the outsized payoffs this year, I guess it seems like the message is if that two- to three-year, two- to four-year window on these loans funding up, I guess, is that suggests that 2021 could see a healthy decline in the payoffs given that normal cycle. Is that the right way to think about it?
I think that's the most logical interpretation of the data. 2018 was $4.74 billion in originations. That is going to be a 2020, mostly 2021, lesser extent 2022 set of payoffs. That compares to the 2017 $9.1 billion. When we originated the 2016 and 2017 large numbers, we knew they were going to pay off at some point in time. The hope was we'd be able to continue to find an ever-growing world of opportunities and keep outrunning that payoff wave.
Obviously, when construction and development activity slowed down and competition got very aggressive in 2018 and we originated less volume, we began to understand as time has gone on that the differential in originations payoff waves was going to create what happened to us in 2019 and will probably happen in 2020, and that's we'll have some pretty slow growth years till we get through that. I do think you're right, that we ought to have less payoff headwinds in 2021 and 2022 than we've had in 2019 and 2020.
Got you. Okay. I appreciate all the color. Thanks.
All right. Thank you.
Our next question comes from the line of Stephen Scouten with Piper Sandler. Your line is now open.
I'm sorry, I didn't have any further questions. Apologies.
All right. Thank you.
With that, I'm showing no further questions in queue. I'd like to turn the call back to Mr. Gleason for closing remarks.
All right. Well, thank you all very much for joining the call today. We greatly appreciate your time and attention. We look forward to talking with you again in about 90 days. Have a great first quarter. Thank you.
Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect.