Bank Ozk (OZK) 2004 Q3 法說會逐字稿

At this time, I would like to welcome everyone to the Bank of the Ozarks third-quarter earnings release conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks there will be a question and answer period. (OPERATOR INSTRUCTIONS). Thank you. I would now like to turn the conference over to Susan Blair. Please go ahead.

Thank you. Good morning. I'm Susan Blair, Executive Vice President in charge of Investor Relations for Bank of the Ozarks. The purpose of this call is to discuss the Company's third-quarter earnings press release issued after the close of business yesterday, our results for the third quarter of 2004, and our outlook for the remainder of the year. Our goal is to make this call as useful as possible in understanding our recent operating results and the future plans, prospects, goals and expectations of Bank of the Ozarks.

To that end, we will make certain forward-looking statements about our plans, goals and expectations of future events, including -- statements about economic and competitive conditions; our goals and expectations for revenue growth; net income; earnings per share; net interest margin, including the effects of the recent purchase of additional bank-owned life insurance; net interest income; non-interest income, including service charge and mortgage lending income; non-interest expense; our efficiency ratio; asset quality; non-performing loans and leases; non-performing assets; net charge-offs; pass-through loans and leases; interest rate sensitivity, including the effects of possible interest rate changes on our net interest margin and net interest income; future growth and expansion, including plans for opening new offices; opportunities and goals for market share growth and loans, lease and deposit growth.

You should understand that our actual results may differ materially from those projected in any forward-looking statements due to a number of risks and uncertainties, some of which we will point out during the course of this call. For a list of certain risks associated with our business, you should also refer to the forward-looking information caption of the management's discussion and analysis section of our public reports filed with the SEC. Forward-looking statements made by the Company and its management are based on estimates, projections, beliefs and assumptions of management at the time of such statements, and are not guarantees of future performance. The Company disclaims any obligation to update or revise any forward-looking statement based on the occurrence of future events, the receipt of new information, or otherwise.

Now, let me turn the call over to our Chairman and Chief Executive Officer, George Gleason.

Good morning and thank you for joining our call as we report another record quarter at Bank of the Ozarks. This is our 15th consecutive quarter for both record net income and diluted earnings per share and our 29th quarter of record net income out of the last 31 quarters.

Third-quarter net income was up 26 percent and diluted earnings per share were up 25 percent compared to last year's third quarter. This strong earnings growth was driven by strong balance sheet growth. During the last 12 months, loans and leases have grown 25 percent, deposits 27 percent and total assets 30 percent. This was all internal growth accomplished without acquisitions, and it was also the kind of profitable growth we want, resulting in third-quarter annualized returns on average assets of 1.66 percent and average stockholders equity of 24 percent. All this is a result of the continued successful implementation of our growth and de novo branching strategy which we've now been pursuing for almost 10 years. In fact, next month we will celebrate the 10th anniversary of the opening of the first of our de novo expansion branches.

During the third quarter we enjoyed excellent loan and deposit growth, with loans and deposits growing at annualized rates of 21 percent and 25 percent, respectively. These growth rates were at the middle to upper-end of the guidance range for both loan and deposit growth, which we had stated to be from the high-teens to mid-20s in percentage terms on an annualized basis. While we continued to report excellent loan growth, our third-quarter asset quality ratios once again reflected our strong credit culture. We will discuss several of these ratios in more detail in a few minutes, but suffice it to say that our third-quarter asset quality ratios were among the best that we have reported.

In the third quarter we again invested for the future, sacrificing some short-term earnings to add three new banking offices. We converted our Dallas, Texas loan production office to a banking office and we opened our first Texarkana, Texas office and our third North Little Rock, Arkansas office. These additions give us a total of eight new banking offices added in the first nine months of this year, equaling the record total for new banking offices added in the full year of 2003. We expect to add three more offices in the fourth quarter, giving us an expected record total of 11 new banking offices added in 2004. While these new offices, like all of our new offices, start out with more expenses than revenue, we believe they are important elements of our plans for continued growth and future profitability. We feel that in the third quarter, we once again maintained a good balance between achieving our short-term earnings goals and our long-term growth objectives. For those of you who may be trying to keep count, as of today we have 46 Arkansas banking offices, three Texas banking offices, and one loan production office which is located in Charlotte, North Carolina.

On September 28th, as we've previously reported, we issue $15 million of new trust-preferred securities which will provide additional capital to help support our expected future growth. In the remainder of today's call, I would like to go into some more detail regarding some of the highlights of the third quarter and some of our expectations for the remainder of the year.

First, let's talk a bit more about loan and deposit growth. I've already discussed the growth rates for the quarter and the last 12 months, and those of you who follow our company know that we have traditionally achieved strong loan and deposit growth rates and that we view Bank of the Ozarks as a growth company. Our growth comes from adding new customers every day, and we were keenly focused on that throughout the Company. The third quarter was another very good quarter for adding both loan and deposit customers. This growth was broad-based, with the majority of our offices contributing nicely to loan or deposit growth, or both. And that's the primary reason we saw third-quarter annualized growth rates of 21 percent for loans and 25 percent for deposits.

While our loan and deposit growth percentages will vary from quarter to quarter, we continue to expect our loan and deposit growth rates for the upcoming quarters to average from the high-teens to the mid-20s in percentage terms. We've had 10 years of great results with our growth and de novo branching strategy, and we believe our best opportunities still lie ahead.

Now let's talk about interest margin. Our third-quarter net interest margin was 4.47 percent, up 4 basis points from 4.43 percent in the third quarter. This was very consistent with our recent guidance reiterated in the past three quarters, in which we stated that we were optimistic about our prospects for a fairly stable net interest margin in 2004. If you look at the last five quarters, our net interest margin has varied only 5 basis points, from 4.48 percent in last year's third quarter to 4.45 percent, back to 4.48, then to 4.43 percent and now back to 4.47 percent in the quarter just ended. We continue to believe that we are probably very close to a neutral interest rate risk position, and we do not expect our net interest margin ratio to move up or down significantly if, as expected, there are additional Fed rate increases in the next few quarters.

On October 1, just a few days ago, we undertook one action that will modestly affect our net interest margin in the coming quarters. On that day, we purchased an additional $18 million of bank-owned life insurance. Of course, the expected increases in cash surrender value from these policies will be recognized as non-interest income. We expect this purchase to result in roughly $216,000 of additional tax-free non-interest income in the fourth quarter. If we had not purchased this Boley (ph), these funds most likely would have been invested in interest-earning assets. Shifting these funds from interest-earning assets to a non-interest income producing asset would have the effect of reducing our net interest margin on a going-forward basis. Assuming that net interest margin doesn't otherwise move significantly up or down in the fourth quarter, the effect of the Boley purchase will be to reduce our net interest margin in the quarter by approximately 7 basis points. While this transaction will reduce our net interest margin a bit, we expect the reduction to be substantially offset by improved non-interest income.

The quarter just ended was our 14th consecutive quarter of record net interest income. And with our cautiously optimistic outlook for a fairly stable net interest margin other than adjusting for the effect of the Boley purchase, and our expectations for continued good earning asset growth, primarily in the form of loans, we believe that prospects are favorable for continuing to achieve improvements in net interest income in the fourth quarter and into next year.

Now let's turn to non-interest income. Service charges on deposits are our largest source of non-interest income. The third quarter was our sixth consecutive quarter of record income from service charges on deposit accounts. While there may be some volatility from quarter to quarter, we expect the service charge income will continue to grow over time as a result of growth in our number of core deposit customers.

Our second-largest category of non-interest income is mortgage income. Of course, third-quarter mortgage income was well below last year's record level achieved during the refi boom, but we were reasonably pleased with our third-quarter results which were within the range that we have experienced over the three preceding quarters. While our mortgage pipeline for October looks good, we know that our mortgage business is usually affected by seasonal factors -- specifically, home purchase activity in many of our markets is usually better in the second and third quarters than it is in the first and fourth quarters. Because of this seasonality, I would not be surprised to see a small downturn in our mortgage volume during the holidays. We have a strong mortgage origination platform, and our long-term goal continues to be to increase our market share in this business over time, even though we realize this business will be volatile due to both seasonal and cyclical factors.

Now let me make a few comments regarding non-interest expense. Non-interest expense was up 13 percent in the third quarter compared to the third quarter of 2003. Our total revenue was up 17 percent in the third quarter compared to the third quarter of 2003, allowing us to achieve our goal of growing revenue at a faster rate than overhead expenses. This allowed us to improve our efficiency ratio to 46.14 percent in the third quarter, which is the second-best efficiency ratio that we have reported as a public company. While this ratio will vary from quarter to quarter, one of our most important goals is to continue to improve our efficiency ratio over time.

One factor increasing non-interest expense in the third quarter was the addition of a number of new bankers. While we've continued to hire tellers and CSRs, branch managers and others, to staff our new offices and continue our growth and expansion, we have also added a number of executive officers and senior-level lenders and managers both to enhance our growth and to manage our growth.

For example, in July, we hired Doug Parker as an Executive Vice President. Doug has a very successful track record in real estate lending. His experience and relationships should result in meaningful loan production for our bank. Also in July we hired Brad Payne (ph) as Russellville President and Blake Tarpley and Angie Smith as Senior Vice Presidents in our Russellville office. The addition of these three experienced and talented lenders with substantial local market knowledge gives us much better prospects for growth in the Russellville market in the coming years.

Brian Cheshire (ph) was hired in July as General Counsel. Brian has an impressive record of handling litigation in commercial matters. He will oversee and coordinate legal matters for our bank and holding company, and his addition should provide needed support to our management team as we continue to grow. In August, we hired Rick Wisdom as President of our new Southwest division in Texarkana. Rick is a Texarkana native with 22 years of banking experience in that market, having most recently served as President of a large regional competitor. We expect his extensive knowledge and local experience and expertise to enable us to become a significant competitor in Texarkana.

In September we hired Gene Holman as President of our mortgage division. Gene is a 30-year veteran of the real estate and mortgage banking business in Central Arkansas. He joined us from a large regional competitor where he had recently overseen mortgage banking operations for them in Arkansas, Texas and parts of Tennessee. The addition of Gene Holman allows Jeanne Earhardt (ph), who founded and built our mortgage operation, to assume her new role as Chairman of the Loan Committee and Senior Lender. Jeanne will continue to help manage our growing lending operations, allowing us, hopefully, to continue to be responsive to borrowers while maintaining disciplined credit and risk management practices.

This group is an unusually large group of senior level hires for a single quarter, and we're very pleased with this recruiting class and excited about their prospects. It requires high-performing lenders to produce high levels of financial performance. We have a very disciplined credit culture and high expectations for our lenders and managers in regard to loan volume, loan pricing, and asset quality. Some lenders find these standards and expectations to be a bit challenging. While most of our new hires fit well, occasionally we lose one. For example, during the last quarter we hired an individual with an excellent resume and good references as President of our emerging North Carolina operation. As it turned out this was not a good fit and this gentleman recently resigned. We will continue in that market our efforts to identify the right leader to head our North Carolina operations.

And speaking of high performance, I want to commend our lenders and credit personnel for another excellent set of asset quality ratios in the third quarter. Our September 30 ratio of non-performing loans and leases as a percentage of total loans and leases was 27 basis points. Our ratio of non-performing assets as a percent of total assets was 23 basis points. Our annualized net charge-off ratio for the third quarter was 10 basis points, and our ratio of loans and leases past due 30 days or more, including past due non-accrual loans and leases as a percentage of total loans and leases, was 46 basis points at September 30.

When you look at these numbers in conjunction with our favorable loan growth rates over the last decade, and our excellent net interest margin, we think our loan and credit team is doing a pretty darn good job of underwriting pricing and growing the portfolio. We ask our lenders to hit some really tough goals for volume, quality and pricing, and once again they came through in a very big way in the past quarter. And I want to publicly recognize them and thank them for their contribution to our results.

In summation, let me say that we believe the third-quarter results are results on which we can build in the upcoming quarters. As I have stated many times, our goal is to continue to improve net income each quarter. This of course means reporting record income each quarter compared to the preceding quarter. We have achieved this goal in 29 of the past 31 quarters, including the last five in a row. And of course, with each success the bar keeps getting higher. But from where we stand today, we believe this is a reasonable goal for the upcoming quarter.

At this time, we will entertain questions. Let me once again ask Shatina (ph) to remind our listeners how to queue in for questions.

(OPERATOR INSTRUCTIONS). John Rodis, Stifel Nicolaus.

Good morning, George. Congratulations on another good quarter. Just a couple of questions. First off, in the mortgage banking business, can you tell me I guess what percent of that business was refi-related?

Yes I can, John. For the quarter purchase activity accounted for 64 percent of our volume and refi activity accounted for 36 percent. So very close to 2/3 1/3 allocation there. (technical difficulty) lower refi percentage than we've seen in the previous two quarters when it was 45 and 43 percent in the second and first quarters, respectively.

But that will probably be something that you'll see going forward to, though?

I would -- you're always going to have some level of refinance activity. If rates push higher it could go down a little further. But you know, probably in that high 20 to 30s is probably a pretty normalized level.

One more question if I may. You talked about your investment in Boley during the -- or I guess on October 1st, so in the fourth quarter technically. Can you just talk a little bit about kind of your process of what you go through to decide to make that investment of 18 million versus saying investing it in other earning assets or loans?

As you know, our regulatory guidelines allow us to purchase bank-owned life insurance basically to defray the cost of employee benefit programs. And there are limitations and guidelines on how much you can purchase, and we are within those guidelines. The cost of our existing benefit programs fully allowed us to purchase this additional Boley. And basically, John, we look at it on this basis -- since we can use this Boley to differ -- defray the cost of those benefit programs, we look at the income generated from the Boley and how effectively it defrays those costs, versus the income on alternative investment. And to get a comparable yield, say, on a municipal security since the accumulated cash surrender value on the Boley is tax-free, a muni is probably the best comp. To get a comparable yield on a muni, we would have to take probably substantially more interest rate risk if you would than you take on a Boley. So we believe that was an advantageous purchase on a relative basis. We purchased only highly rated companies, AA to AAA type ratings on the companies, diversified our purchase between several companies to diversify the credit risk on those policies.

Barry McCarver, Stephens Inc.

Good morning, George, and congrats. Another great quarter. It's becoming old habit.

We ought to continue it.

A couple of questions, George. First off, can you talk about what your variable rate loans did in the quarter? I think you've had a goal to improve that each quarter.

Barry, that's a great question. We didn't improve it this quarter. In fact, it's the first quarter in many quarters that we actually went backwards. And I'm still drilling down in the data to try to figure out what happened. But variable rate loans at the end of the preceding quarter, at June 30, were 37.5 percent of our portfolio. And you recall we were very pleased with that, because that was almost a 400 basis point increase from the prior quarter. But this quarter we went backwards, and variable rate loans were at 35.2 percent of the portfolio. So we actually went backwards 2.3 percent there, 230 basis points compared to the prior quarter. And again, I don't have an explanation for that. We're still drilling down into the data. There's nothing at a high level that answers that question for us.

What is encouraging to us, though, is that notwithstanding that, the repricing attributes of our portfolio didn't change much. If you add in fixed-rate loans that mature and the principal cash flow from the regular scheduled payments on fixed-rate loans to really get how much of the portfolio reprices, in the first year, now 58 percent of total loan portfolio will reprice. And that's actually up even though the variable piece is down. I guess the loans maturing and the cash flow coming off of the fixed part of the portfolio in that one-year timeframe is up more than the decline in variable rate loans. So we went from 57 percent repricing in one year to 58 percent. In two years that number held steady at 71 percent, and in the three-year time frame we went down from 83 percent of the portfolio repricing to 82 percent repricing. So slightly better in the one-year, same in the two-year, slightly less favorable in the three-year timeframe. And all those numbers either were the same or moved 1 percent. So despite the decline in variable rate loans, the interest sensitivity of the portfolio really did not change in an appreciable manner in the quarter. It continues to be our goal to increase that percentage of variable rate loans, and I do think we will be successful in doing that in the future, and hopefully in this quarter.

What about commercial loan growth? Did you see any pickup in that?

Barry, as you know, the majority of our portfolio is real estate related. And I would say generally across the portfolio my sense is -- and again, I haven't drilled down into the details out on this yet -- but my sense is that the growth was pretty broad-based across the portfolio, consumer, construction and development lending, commercial lending and residential (indiscernible). I'm not aware of any particular change in the recent trends that we have had that have been toward that becoming a slightly more commercial mortgage portfolio over time.

George, last question. If I remember correctly after the second quarter, I thought you said you were expecting to open one new branch during the third quarter in addition to the conversion of the branch into the Wal-Mart store. And now we ended up with three. Is something changed? Because you still say three in the fourth quarter as well. So it sounds like you're going to do more this year then what you had originally planned.

Two things occurred that did change our plans. One is the Dallas, Texas loan production office -- that has been a very successful lending operation for us -- made the request that they be allowed to take deposits in their LPO facility. So we did make a decision to file a branch application to turn that LPO facility into a temporary banking office, because they had deposits opportunities with some of their commercial lending customers there that they were having to forego because of our lack of powers to handle deposits in that branch. So we made that change, and that was one addition. The other addition is, as I mentioned in my prepared remarks, we hired Rick Wisdom as Southwest division President in Texarkana, Texas. And the Buoy (ph) County, Texas, Miller County, Arkansas markets are markets we had planned to enter in the future. But Rick became a guy on our radar screen. We had some meetings with him. We hired him and accelerated our plans to enter that market because of the availability of what we think is an outstanding leader for that market. So that did put a couple on the radar screen that were not on the radar screen. And assuming we do get the three open that are under construction and should open this quarter, that would put us at 11 for the year. That's a pretty high tally.

Peyton Green, FTN Midwest Research.

Kind of looking forward away from the fourth quarter and into next year, what do you expect to do on the branch side? And then in terms of the growth that you have booked year-to-date in '04, how would you characterize the growth from your de novo versus your existing offices?

Peyton, let me answer those questions in reverse order. As I travel around and talk with investors or present at conferences, I frequently get asked to give a same-store sales number. And we don't actually look at it and calculate it on that basis. We have a growth target for every office, every year, an expectation and a plan for what we want to do. But, you know, a bank branch is a little bit different than a retail store. You open a retail store, and after the first year, after that grand opening and start-up phase, a retail store pretty much ought to have a baseline set of data that you can calculate from. A bank branch -- hopefully, you get it up and get it profitable somewhere around a year and get a break-even or a larger volume of business in there -- but that branch ought to continue to grow for a long, long time.

So it's a little bit different than a same-store analogy, and hence, I don't know that if we looked at our branches that were open more than a year and gave you a growth rate on those, that that would really give you very much of a meaningful number. So we don't calculate it that way. What I would tell you is is that our growth in the quarter was, as I said in my remarks, fairly broad-based. Most of our offices either had good growth in loans or deposits, or both, in the quarter, and contributed nicely to that growth. Obviously, with this kind of conveyor belt, if you would, of de novo branches, a de novo branch typically that has been open eight years or six or four is going to be growing at a progressively slower rate because of the bigger denominator and the lesser growth opportunities than one that has been open for two years. And that pattern is pretty well established, although there are occasional exceptions and deviations from that in our company.

The other question you asked is what is our plan for de novo branch expansion next year. And I would tell you that we would expect that a number of new offices -- and it's not in concrete yet, although we have a fairly specific plan, it's still subject to some amendment and change -- but we would expect to open probably something between the level we opened last year as the low-end of the range and the number we've opened this year as the high-end. Of course in '03 we opened eight offices; this year we are expecting to open 11. So our present thought -- and again, this could change an office or two -- but our present thought is that we probably would be looking at somewhere between eight to 11 offices next year.

Just a follow-up to the variable versus fixed in terms of the loan mix. In terms of pricing, do you feel like that you got compensated for the fixed rate activity in the quarter versus maybe wanting to do the floating rate before? But do you think, given now a little bit more subdued interest rate outlook, that it might have been the right thing to do, or is it just something that happened because of customer demand?

Peyton, my sense is it's just something that happened because of customer demand. And yes, my sense also is is that we got compensated reasonably for it. Hopefully we did. Our standard policy, and all of our lenders understand this, that if we do a two-year fixed-rate loan or a three-year fixed-rate loan, or a one or a five -- whatever the term is -- that we've got to get -- depending on how long it's fixed, we've got to get increasingly higher rates over the rate that we would have charged had we done that on a variable-rate basis to compensate us for that interest rate risk. And that if it's a three-year fixed-rate loan, that the interest rate risk is more substantial than a one-year, and hence we ought to get paid proportionately more premium rate over and above the variable rate pricing of that loan, to take that additional interest rate risk. Our lenders understand that. They price accordingly. And I don't think I can tell you that we always get paid for the interest rate risk we take. Probably no banker does, because occasionally you get pushed into deals for competitive reasons that don't make pure mathematical sense from an interest rate risk. But generally I would tell you that I think we got -- we probably got compensated for that bit of additional interest rate risk.

Then as a follow-up, the competitive conditions on the deposit side -- how do you feel about that now that I guess the initial shock of interest rates going up has kind of subsided? And also on the loan side, are you seeing any change in competition there?

I would say on the loan side first, we've not seen much change there. It continues to be as it has been for the last couple of years, very competitive on the loan side. And we do frequently encounter one competitor or two competitors in specific markets that are doing in specific transactions very aggressive sort of one-off pricing deals that strike us as being poorly priced or not adequately priced for the interest rate or credit risk. On the deposit side, we have been a little surprised in the competitive response. We modeled this competitive response out very, very specifically based on markets and products and competition. And what we have seen in response to the three Fed rate increases is that our competitors have not moved their administered product rates, their NOW accounts, checking accounts, money market and savings accounts much at all. And we have seen more movement than we expected on the CD side. Now when you blend those two things together -- and we have done that and we have modified our response slightly to match up with the competitive pressures -- but when you blend those things together, we had just about the exact net interest cost that we expected. We just spent more on CDs and less on adjustments of administered rates then we had expected to, but the net of that was almost exactly what we had modeled and projected. So it didn't move our numbers at all from where we expected them to be, but the components of it were a little bit different.

That also -- the product of that is it skewed our deposit mix a little bit. In the third quarter, our non-CD deposits, which had been 46.2 percent of deposits in the prior quarter, dropped to 44.8. And our CD piece went from 53.8 to 55.2. And that's simply a result of the fact that we had a number of competitors very aggressive on the CD side in the market, and I'm sure they cannibalized some of their own non-CD deposits as a result of that, just as we cannibalized some of ours as we responded to their CD rates. So it did shift the mix a little bit. It did shift the cost of our CDs, but the net effect was that we ended up where we thought we would as far as net interest expense.

Okay. But I mean, wouldn't you want to have a little bit more CD mix if you have a little bit more fixed-rate load exposure? Wouldn't it kind of balance out on the interest rate risk side?

Our policy and our goal would say we want to keep that CD mix and non-CD ratio somewhere in that 45-50 to 50-55 range -- as close to 50-50 as possible. But given the prospects for some more rate increases, even though expectations have moderated somewhat on that, we really are not too sad about having a little more duration on the deposit side. But at the same time, we would like to get back to that kind of 45 to 50 mix on the non-CD's and 50 to 55 on the CDs. And at 55.2, we are a little bit outside kind of the far edge of that range. So I hope that will work back the other way, but part of that is going to be determined by whether our competitors sort of balance their rate-adjusting strategies on a go-forward basis or continue to be very CD-reliant as far as adjusting rates and trying to attract deposits.

Last two questions and I'll drop off. What is the Boley investment now in terms of your balance sheet?

We purchased 18 million and we had -- a couple of years ago we purchased 20 million. It's grown to about 22 million. So the total investment now is about $40 million.

Do you have any sense what your interest rate risk up 100, up 200 was at September 30? And that's it for me. Thank you.

We do not have that data, and it will be several weeks. We typically finish that simulation model literally about the week that we file the Q. So it should be in our Q as it always has been. Our general sense without actually running the numbers, Peyton, is that the interest rate risk results are not materially different than they have been the last couple of quarters, which is plus or minus -- very little deviation from a neutral position.

Chris Kelly, Commissum Capital.

Good morning, George. Quick question. When you filed the 8-K about the trust-preferred, you also dropped in there a little paragraph about the entry into a material agreement. I'm just wondering was there any income statement -- the numbers aren't that big -- but was there any income statement impact in the third quarter, or will we see any in the fourth quarter that pertains to something that almost reads to me like a recovery?

No. There's no income statement impact at all from the transaction. For those of you who may not be familiar, we had one of our executive officers and another officer in the Company that wanted to buy an asset that we had that was in problem loan status for a personal investment. And they made the Bank an offer to acquire this asset, and actually acquired the loan, which we assigned the note and the mortgage and so forth to them. We handled this transaction independently. The matter was turned over to our president. He obtained independent appraisals on the two underlying pieces of collateral, did an evaluation of our remaining cost to collect, foreclose and liquidate that, and determined that the offer that we had from these two officers was a satisfactory offer. We accepted it and assigned the note and collateral package to them. It was handled in a very arm's-length sort of way and was done exactly like we would do it with a third party. There was no fourth quarter impact at all. The transaction was fully closed in the third quarter, and there was no appreciable -- I mean, it was a few dollars, I think, Chris, one way or the other -- impact from the actual sale of the asset to them. Now, we had taken a couple of months earlier a small charge-off on the deal. And that was, of course, reflected in the quarter's 10 basis point charge-off number. But that had occurred a couple of months earlier in my normal monthly, and was a result of discussions of that asset and estimations of value of the underlying collateral in my normal monthly past-due calls with our lenders.

Okay. One last thing. I don't know if you've got this number at your fingertips or not, but like most folks, you swung down in the unrealized loss on the securities portfolio in June. I assume you galloped back a good bit of that this quarter. Do you have where that ended the quarter?

Yes. On a pre-tax basis, the unrealized loss in the securities portfolio was $2 million at the end of the third quarter. And that was a sizable swing from where we were at June 30, and honestly, quite unexpected on our part. We felt that the yield curve would flatten; we thought the curve would flatten by the front end rising and the five-year and 10-year sectors rising less. We did not envision the substantial rally in five and 10-year parts of the curve that we actually had in the quarter.

I think you've got company there. I appreciate it.

(OPERATOR INSTRUCTIONS). Scott Alaniz, Sandler O'Neill.

A question for you. On the securities portfolio, you have an above-average above-peer group yield. And that also increased sequentially. If I look at the taxable portfolio, increased from 487 to 522. Could you talk a little bit -- and the size of that actually increased as well. Could you talk a little bit about what types of products and the duration of those products that you're adding to the securities portfolio please?

Yes, would be happy to do so. Let me first talk about the increase in the size of the portfolio and tell you that the size of the portfolio as a percentage of earning assets was a little higher at the end of the quarter than we intended and expected. At September 30, our securities portfolio accounted for 28.71 percent of earning assets. Our goal is for that to be 22 to 28 percent, and we got out of the top side of that. And the basic reason for that was when we calculated and budgeted our securities purchases at the beginning of the quarter, we had three things that turned out a little bit differently than we expected. And one of those is -- and I mentioned in response to Chris's question -- we did not imagine this wide swing back in the mark-to-market adjustment. And that alone took us from -- if we had pro formad (ph) the June 30 marked-to-market adjustment in the September 30 numbers, that ratio would've been 2801. So that would have us just right at the top of our range. And secondly, we didn't have quite as much loan growth for the quarter. We were a few million dollars short on the loan growth number that we had modeled. And thirdly, we found the opportunity to buy some Arkansas munis in the quarter. And the Arkansas muni market, as you probably are aware, is sort of a chunky market. And sometimes there's no supply out there and sometimes there is. And we found a few deals that gave us a chance late in the quarter to buy some Arkansas munis that we wanted to own for portfolio. And we just hadn't budgeted that, so we went ahead and plugged those in. So that pushed us a little over our 28 percent target ratio, and we have adjusted our fourth quarter purchases downward and have very nominal levels of security purchases planned for the fourth quarter, because we want to get back under that 28 percent ratio this quarter.

The securities that we purchased in the third quarter, other than the munis we purchased, were primarily CMO mortgage-backed type product. Of course, AAA-rated credit product that does have some extension and risk to it. And that is one of the attributes that gives us that above-average yield on that. We feel like as we've modeled and studied each of those securities, that we get a good premium yield that more than compensates us for that risk in that CMO ode mortgage-backed portfolio, the extension risk of it. As of a couple of days ago, actually the fifth of this month, the average life of our entire securities portfolio is modeled by our service provider that does our portfolio accounting. The average life was 3.4 years. The duration was 2.8 years on the portfolio. And that average life is down from numbers that I would have given you, and I think did give you, in the conference call in July. And that's just because of the flattening yield curve, the expected prepayment speeds, and average life on that mortgage-backed portfolio is shortened a year or two, depending on whose model you use over the quarter because of the flatter year curve.

Second question. Non-performing assets. Could you perhaps provide us a little color as to the composition, and if there are any particularly large credits, if you will, that are in that category?

Obviously, there are not any real large credits in there, Scott, since they comprise only 27 basis points alone and 23 (multiple speakers)

That's true.

There really is not much chunkiness to that. I think the highest single asset in there -- the total non-accrual loan is 2.9 million. The largest single asset that I believe that is in either OREO (ph) or loans non-performing is approximately $4 million, and we've got a contract to sell that that should close any day. So there's nothing there of a concentration nature that is an issue. It's just kind of your normal, routine collection of things going through the collection process.

I see. Lastly, the timing of the fourth quarter branch openings. Are these three openings expected earlier in the quarter, later in the quarter?

Scott, once they may actually happen like we huddle them about. If they go as planned now, we'll have one in October, one in November and one in December. Which is how we normally plan these. It seems that despite that good planning, invariably, in most cases we end up with all of them in one month. Because this one got delayed a month because of some construction problem, and this one got accelerated a month because the contractor got ahead of schedule and they end up clumped together. We plan to open them kind of on an orderly, measured pace, to use the Feds term. And sometimes they just end up in chunks. But it looks like now we're going to open one a month this quarter.

At this time there are no further questions.

There being no further questions, let me thank you guys for joining the call today and tell you we look forward to talking with you again in January, and appreciate your interest in our company. Thanks very much. Have a good day.

Thank you. This now concludes today's Bank of the Ozarks third-quarter earnings release conference call. You may now disconnect.