Zions Bancorporation N.A (ZION) — Q2 2017 Earnings Call Transcript

Thank you, Latif, and good evening. We welcome you to this conference call to discuss our 2017 second quarter earnings. To begin, we will hear prepared remarks from Harris Simmons, Chairman and Chief Executive Officer; and Paul Burdiss, Chief Financial Officer. Additional executives with us in the room today include Scott McLean, President and Chief Operating Officer; Ed Schreiber, Chief Risk Officer; Michael Morris, Chief Credit Officer; as well as other Zions executives who are available to address your questions during the question-and-answer section. I would like to remind you that during this call, we will be making forward-looking statements, although actual results may differ materially. We encourage you to review the disclaimer in the press release or the slide deck dealing with forward-looking information which applies equally to statements made in this call. A copy of the full earnings release as well as a supplemental slide deck are available at zionsbancorporation.com, and we will be referring to these slides during the call. The earnings release, the related slide presentation, and this earnings call contain several references to non-GAAP measures, including preprovision net revenue and the efficiency ratio, which are common industry terms used by investors and financial services analysts. Certain of these non-GAAP measures are key inputs into Zions' management compensation and are used in Zions' strategic goals that have been and may continue to be articulated to investors. Therefore, the use of such non-GAAP measures is believed by management to be of substantial interest to the consumers of these financial disclosures and are used prominently throughout the disclosures. A full reconciliation of the difference between such measures and the GAAP financials is provided within the published documents, and participants are encouraged to carefully review this reconciliation. We intend to limit the length of this call to 1 hour. During the question-and-answer section of the call, we ask you to limit your questions to 1 primary and 1 related follow-up question to enable other participants to ask questions. With that, I will now turn the time over to our Chairman and CEO, Harris Simmons.

Thanks very much, James, and we want to welcome all of you to our call today to discuss our second quarter results. On Slide 3 are some highlights for the quarter. We were pleased with the strength of the loan growth and the sources of that growth, notably C&I loans, excluding oil and gas loans, consumer loans, and there was less dependence on commercial real estate loans than in years past. We were pleased to contain the cost of deposits, and the combination of those two factors led to a healthy increase in sustainable net interest income. Finally, we're very pleased with the trends in credit quality, particularly in the oil and gas space, which has been a substantial portion of our problem credits in recent quarters. I'll cover the rest of the key indicators listed on this page as we move through the presentation. On Slide 4, our adjusted preprovision net revenue reflects steady improvement, up 27% year-over-year and about 19% if one excludes the interest income recovered on four larger previously charged-off loans that we call out as unlikely to be sustained. Some of the growth is due to the benefits of rising interest rates and securities purchases, which we do not expect to be as significant of a contributor to growth over the foreseeable future, but we do expect continued solid loan growth and customer-related fee income growth that we expect will contribute to further positive operating leverage. Turning to Slide 5. We posted an efficiency ratio of 59.8% in the second quarter. Excluding the interest recoveries described throughout the release, the ratio would be closer to 61%. Because of seasonal factors in both revenue and expense, it's more appropriate to compare the efficiency ratio to the year-ago quarter. As shown in the chart on the left, our goal for 2017 is to achieve an efficiency ratio in the low 60s range while holding adjusted noninterest expense growth to between 2% and 3% from the 2016 actual results of $1,580,000,000. As we have highlighted many times before, we're achieving all of this while making very substantial technology investments in new core systems and related projects. We were very pleased with the successful implementation of the consumer loan module of the core system, which took place over the Memorial Day weekend. And I really want to complement hundreds of employees and particularly the management team for that project on a very successful systems conversion for our consumer loan portion of this project that took place several weeks ago. We expect that these investments will ultimately make the company more efficient and reduce risk, and beyond these technology investments, we're investing in the hiring of relationship managers and similar positions that we expect will enhance our revenue growth. In the back office, we continue to work hard to identify additional opportunities to become more efficient. Moving to Slide 6. We experienced softer loan growth over the last years than was desired, but as we said in conference appearances in recent weeks, we experienced a solid second quarter, up nearly 9% on an annualized basis. One of the strengths this past quarter has been on 1- to 4-family loan area, which experienced an 11% annualized rate of growth and had a similar growth rate over the prior year's level. We also experienced strength in commercial and industrial and owner-occupied loans. We were not dependent upon commercial real estate for growth in the quarter, although we continue to originate new loans for customers. We're adhering to our internal concentration limits, and we expect only moderate growth in commercial real estate as a result. Loan growth outpaced deposit growth in the quarter; with a moderate deposit attrition period, we increased wholesale funding and utilized cash and securities to fund the loan growth. On a year-over-year basis, period-end deposits increased 4.2%, and average deposits increased 4.8%. Slide 7 depicts the overall credit quality metrics of our loan portfolio. We're encouraged with a meaningful improvement in classified loans, nonperforming assets, and net charge-offs. Much of the improvement came from the oil and gas portfolio, and we remain optimistic that we'll continue to see further favorable changes to oil and gas credit metrics as we continue to see strengthening revenue and cash flows from many of these companies. Slide 8 depicts two key profitability metrics: return on assets and return on tangible common equity. Return on assets increased to 103 basis points in the second quarter. And excluding the items listed in the bullet, the ROA was in the mid-90s, up from the mid-70s a year ago. Return on tangible common equity increased to 10.2% from 6.3% a year ago. And excluding interest recoveries on four loans, it would have been in the low 9% area. I'd note that we achieved that with one of the strongest equity positions among regional banks. We're encouraged by the continued improvement and remain focused on achieving highly competitive returns on our balance sheet, both assets and equity, relative to peers. Before I turn the time over to Paul to discuss the financial results further, I'm pleased to report that our capital plan for the next four quarters triples the common dividend rate and increases the dollars allocated to share repurchase by more than 250% relative to the prior four quarters. Given the expected balance sheet growth, we expect we'll see some moderate increase in the leverage of the balance sheet, which should also help improve our return on equity. With that overview, I'm going to turn the call over to Paul Burdiss to further review the financial results.

Thank you, Harris, and good evening, everyone. I'll begin on Slide 9. For the second quarter of 2017, Zions reported net earnings applicable to common shareholders of $154 million or $0.73 per share, which is up from $0.44 per share in the year-ago period. There are some items within the second quarter 2017 result that are operating but generally infrequent and those include interest recoveries of $60 million on four loans that had previously been charged off; securities gains of $2 million; income tax benefits that are not expected to recur in a material amount, which were about $4 million in the second quarter; and accelerated recognition of preferred stock issuance costs of $2 million due to the redemption of preferred shares in the second quarter. These items amount to about $13 million in after-tax dollars or about $0.06 per share. Let me make a few comments about revenue. Approximately 80% of our revenue comes from net interest income. Slide 10 depicts the recent trend in net interest income, which continued to demonstrate growth in the second quarter. On a year-over-year basis, quarterly net interest income was up $71 million or 14%, and about $55 million or 10% if one excludes the aforementioned loan recovery income. The most significant factor in the year-over-year increase in net interest income is the $5.9 billion period-end and $6.7 billion average balance increase of the investment securities portfolio. Slide 11 shows the growth in the period-end balances. If I extend the time period back just a bit further, since the fourth quarter of 2015, the growth of securities has resulted in about $185 million of increased interest income annually. Before I move on to the next slide, I will reiterate that we continue to exercise caution with respect to duration extension risk, and combined with a relatively flat yield curve, we utilized some of the cash flow from the securities portfolio in the second quarter to fund our excellent loan growth. Slide 12 is a graphical representation of our loan growth by type, relative to the year-ago period. The size of the circles represents the relative size of the loan portfolios. Total growth, including the effects of a declining oil and gas portfolio and the national real estate portfolio, was 2.8%. We experienced 4.2% year-over-year growth in loans outside of the oil and gas portfolio, and we now expect oil and gas loans to stabilize generally from here, providing some improvements to the overall growth profile over the next several quarters. The key takeaway from this chart is the relatively balanced growth for most of the circles, i.e., loan portfolios. Commercial and industrial, owner-occupied, construction, and home equity loans all increased in the mid-single-digit range while we experienced solid growth in the 1-4 family loan portfolio, offset by weakness in oil and gas and national real estate loans. Excluding national real estate loans, commercial real estate loans were relatively stable, up only $27 million over the prior year. As we discussed over the past several quarters, our internal commitment to manage our portfolio of concentration limits is the reason for the general stability in the commercial real estate portfolio. Shown in the bottom is our expectation for loan growth by product type. Turning to the rate component of net interest income on Slide 13, this slide breaks down key components of our net interest margin. The top line is the loan yield, which increased 24 basis points from the prior quarter to 4.38%. But as noted in the press release, about 15 basis points was attributable to key recoveries of interest income. The remainder of the increase was primarily attributable to the increase in short-term benchmark interest rates. The securities portfolio yield decreased 4 basis points this quarter compared to last quarter, which is due largely to higher premium amortization on our SBA portfolio within the securities portfolio. Our cost of total deposits increased only slightly relative to the prior quarter. While we're monitoring the competitive landscape and will act accordingly, we're not seeing systemic pressure on deposit pricing. The cost of interest-bearing funds increased somewhat compared to the prior quarter due to increased wholesale borrowings, which were used to fund our strong loan growth in the second quarter. On the bottom right side of the slide is a table of the various indices to which our loans are indexed, with the first column of numbers showing the gross percentage of loans linked to the respective indices, and the second column of numbers showing the offsetting effects from floors and swaps, while the third column represents the net effect. Compared to the prior quarter, there is not a significant difference, although some loans have moved off of their floors at this point, and beginning in July, the swap portfolio has begun to experience some limited attrition. On Slide 14, we have adopted our new model, which is a single-point estimate of the effect on net interest income in a rate environment that is 200 basis points higher than the current level. We have also shown the results of the model we had been using for the last several years for comparative purposes. The primary difference between the models is a refinement in the new model, regarding how much of the deposit base is treated as core versus transitory. Turning to Slide 15 and noninterest income, total noninterest income was $132 million, which was up from $126 million a year ago. Customer-related fees, shown on the slide, increased to $121 million from $118 million a year ago and were up from $115 million in the prior quarter. We made good progress on card-related income, up $2 million from the prior quarter, loan fee income which is up about $1.5 million on good progress with our new loan syndication platform, as well as good performance on Wealth Management and on customer interest rate hedging activity. Turning to management activity, it was relatively stable from the prior quarter. As noted in the press release, non-customer related activity, such as securities gains and dividends from small business investment company investments, contributed to the bottom line, although not as much as in the prior quarter. As we continue to focus on customer-related fee income, we're still targeting mid-single-digit annual growth for the full year 2017. Noninterest expense on Slide 16 increased to $405 million from $382 million in the year-ago quarter. If adjusted for items such as severance, provision for unfunded lending commitments, and other similar items, noninterest expense increased to $399 million from $385 million in the year-ago period. Notably, salaries and employee benefits were relatively stable from the year-ago period at $242 million, up only $1 million. Year-over-year, FDIC premiums and the revenue-sharing agreement with the FDIC on loans purchased equaled about $6 million, the former of which is largely due to the temporary surcharge by the FDIC under the Dodd-Frank Act and the latter is linked to stronger revenue performance. Both items are somewhat transitory in nature. Excluding these items, the year-over-year increase in adjusted noninterest expense was approximately 2%. The key changes relative to the prior quarter are listed on the slide, so I won't take any more time here other than to reaffirm our expectation that total noninterest expense is likely to increase between 2% and 3% in 2017 when compared to the 2016 actual results. On Slide 17 is a list of our key objectives for the remainder of 2017 and 2018 and our commitment to shareholders. We're fully committed to continue to achieve positive operating leverage. We have a couple of years behind us at this point in our effort to materially improve profitability and grow earnings. We remain committed to further improvement and simplification of our operational processes. Harris mentioned the foundational technology improvement earlier in his comments today, so I'll move on to capital. When we first rolled out this slide, we indicated that we were going to be targeting much more substantial returns on capital than what could be seen then, and we're tracking well toward those goals. Regarding returns of capital, we indicated that we plan to be more assertive in our capital plan for the Horizontal Capital Review in 2017, and we're pleased that the Federal Reserve did not object to our new capital plan, the key actions of which are outlined on this slide. Finally, Slide 18 depicts our outlook for the next 12 months relative to the most recent quarter. We're maintaining our outlook for loan growth at moderately increasing, which is to be interpreted as an annual rate of growth in the mid-single digits. Using a modified net interest income of $512 million, which excludes the $60 million of interest recoveries on four loans that we have called out as transitory, we expect net interest income to increase moderately over the next 12 months. We're not assuming an additional rate hike in this outlook, and we do not expect further benefit from the rate increase thus far in 2017. Additional increases in short-term rates are expected to improve net interest income. Turning to the provision for credit losses, we posted a net provision for both funded loans and unfunded commitments of $10 million in the second quarter. We said throughout the quarter that we were becoming increasingly optimistic that the credit deterioration from the oil and gas portfolio was turning or had turned the quarter. Despite somewhat lower prices for oil, the health of the companies within our portfolio continues to improve and we're effectively reducing our outlook for credit costs from the April outlook which was about $18 million per quarter down to about $10 million per quarter or maybe slightly higher, with a few million dollar range of imprecision. This also assumes that energy prices do not experience a substantial decline from the current level. We expect that customer-related fees which are defined in our press release and exclude dividend income and securities gains and losses, should increase moderately from the level reported in the first quarter. We currently expect adjusted noninterest expense to increase in 2017 between 2% and 3%, relative to the 2016 reported results. Because our outlook now extends to the first half of 2018, in order to preempt the question about growth in 2018, we believe it could grow in a similar range for that year as well, although we're not planning to loosen up on our efforts to streamline and simplify various processes. One example we've begun to shine a spotlight on is procurement in taking costs out of our supply chain. We're in the first inning of that effort. We don't expect large numbers of physical branch reductions, as digital adoption continues to take hold, but we do expect some fine-tuning of physical branch locations. At the same time, we want to hire strong talent that will drive revenue growth and enhance the value of the franchise, and that is the general rationale for expense levels which may increase slightly. Excluding the adjustments for the 2017 accounting guidance for stock-based compensation, we expect our effective tax rate to be in the 34% to 35% range for the next four quarters, barring any meaningful changes in the tax code. We expect cumulative preferred dividends to be approximately $34 million over the next four quarters and diluted shares may fluctuate due to both share repurchases and the dilutive effect from our outstanding warrants. The dilutive effect of the warrants is predominantly dependent upon the price of our shares which we described in the January earnings report. Please see the appendix of our test slide deck for further sensitivities on the warrant effect. This concludes our prepared remarks. Latif, would you please open the line for questions? Thank you.

You guys had included in your capital return for this year, in your repurchase, you'd included the up to the full CCAR ask amount. I mean, do you anticipate anything that could push you off a trajectory to complete that full amount?

Well, we don't at the moment, but it's obviously something that the Board's going to look at every quarter before they declare it and see if conditions have changed, either deterioration in the general economy or our portfolio, et cetera, but at the moment, we don't foresee anything.

Got it. And then, deposit costs were really well-controlled this quarter. How long do you guys feel like you can keep a lid on that? I mean, obviously, you noted you're not seeing a whole lot yet. We have started to hear from some of your peers that there's a little bit more competition on the commercial side of things, but just be interested in your outlook on that.

Yes, we're seeing a little more competition which I referenced in my comments. We're not seeing systemic pressure, but we're seeing sort of one-off, which we're addressing through our regular processes. You would imagine this is critically important to our interest sensitivity and therefore, net interest margin and income and profitability, so we're watching this very, very closely. But as I said, we're not yet seeing any systemic pressure on deposit pricing.

Paul, I would ask you about the margin versus the NII. As you've been purchasing securities in the last couple of quarters, there wasn't really a substitution, so your balance sheet actually expanded. Is that why your margins didn't kind of expand as much, but yet NII kept the same pace of growth?

Yes, Marty, by my calculation, these are very rough and approximate and probably subject to debate. But the securities portfolio growth that we experienced in the first quarter was a headwind of about 8 basis points or so on the net interest margin. In the second quarter, I calculate that headwind to be 4 to 5 basis points. So that is to say, you correctly point out, the purchase of securities has actually expanded the size of the balance sheet, whereas in the past we were just replacing cash. So therefore the sort of marginal spread of what is being added is lower than that strict cash replacement, which is why you're seeing a little bit of a headwind. Now with all that being said, as I said at the end of the first quarter call, the securities portfolio growth is effectively done. In fact, you saw the securities portfolio, the ending balance was actually down a little bit, even though the average was up. So my expectation going forward is that net interest margin headwind will no longer be there.

And then a follow-on to that. The second stage was the interest-bearing swap as you add to and begin to neutralize the sensitivity through the swaps. Do you still feel like you're going to do that? And where will that show up? In what line item will that be a benefit?

Well, Marty, I can tell you that at the end of June, our swaps portfolio was approximately $1.4 billion with a carry of about 10 basis points. The yield curve has flattened significantly, making swaps less appealing than in the past. However, we are examining what is rolling off and considering strategies to replace those swaps in order to maintain our current asset sensitivity. Unless there is a significant change in the shape of the curve or the risk-return profile for extending duration, I don't expect that we will materially decrease our asset sensitivity from this point onward, all other things being equal.

Just as a follow-up on the balance sheet. So Paul, if you're properly aligned now on the securities book, in the last few quarters you've been utilizing wholesale borrowings and short-term funding to address that GAAP, which you indicated you would. However, this quarter, deposits have also decreased slightly. Can you explain the relationship between deposit growth and short-term borrowings? Is any of the deposit decline associated with indirect data? How are customers reallocating their funds, and is any of it connected to the rate increases we've observed?

Yes. Sure, Ken. It's hard to specifically tie deposit movements to the change in the rates, at this point, because we're still kind of in the early innings, right? Anecdotally, I've heard of cash being used by our businesses for kind of investments and other things, and so I'm not seeing anything systemic that would lead me to think that we have got deposit dollar movement because of changes in rates. That being said, while it's true that our ending balances were down, I do note that our average balances were actually up quarter-over-quarter. We do, because of the nature of our deposits, we do see some volatility on any given day, and so any time period-end balances, for example, you compare any two days and you could see some volatility. So I'm actually watching average balances much more carefully than the ending balances. And I am not seeing a discernible trend that is causing me concern, to think that we're beginning to see deposit runoff. All that being said, deposits, well-priced, relationship-driven deposits are kind of really a big part of the core value of this organization. This is something we watch very, very closely and take very seriously, and we will protect our deposit base.

Understood. Well put. I have a question about credit. You mentioned that recoveries are good and that all non-performing asset categories are looking positive. There’s still that 8% on energy. Additionally, you noted that provisions might be slightly higher than this quarter, which had low losses of 7 and overall credit of 10. Can you discuss your expectations regarding why provisions appear to remain so low despite the lack of underlying credit issues?

Maybe I'll start with that, Scott, if that's okay, and you can jump in. What we mentioned in the first quarter was that energy was expected to improve, and we've actually seen that, with an expectation that the trend will continue. You pointed out the energy reserves, and I want to remind everyone that we're still maintaining over an 8% reserve on that energy portfolio. Although the dollar amount of that reserve has decreased, the portfolio itself has slightly decreased too, which explains the strong reserve ratios. That being said, the rest of the portfolio is performing very well, making it difficult to believe that this can go on indefinitely, considering where we are in the credit cycle. As we assess potential weaknesses, we focus on everything else in the portfolio outside of energy. While we currently have no signs of deterioration, it's certainly a possibility we acknowledge, and we incorporate that into our outlook.

Hi, this is Scott McLean. I would like to add that if energy prices remain stable or improve, we will see continued improvement in that part of the portfolio, which will reduce pressure on the provision. However, as Paul mentioned, we are being cautious about our position in the credit cycle. Additionally, loan growth is strong and helps alleviate some of the provision requirements as well.

I was wondering here, given your constructive comments on loan growth, how should we think about funding it, whether it's securities, deposits, or continued increases in short-term borrowings?

Well, I'm certainly hopeful that we can continue the trend of solid deposit growth that we've seen. That is our first and best kind of source of funds. You correctly point out, we have, over the course of the last quarter, dipped into our cash position which I think on a relative basis, is still very healthy and probably a little higher than it needs to be. And the securities portfolio is also sort of a ready source of funding. Absent all of those, then that's when we would look to the wholesale market. And we looked at that, I would say last, but we've got a lot of availability there. If you look at our debt relative to peers, you'll see that what we're one of the most deposit-funded organizations. And so while we expect to continue to be almost entirely deposit funded, we do have capacity to incrementally add debt to support loan growth.

Okay. And then, what were new money yields here for the second quarter?

Are you talking about loan yields or?

Loan originations for the quarter.

Loan originations, probably about 415 basis points on a yield basis.

The common equity Tier 1 is still at 12.3%. Clearly, the higher buyback of dividend, coupled with loan growth helps to bring that down a bit. But what's the right level for that ratio longer term? And how quickly do you think you can get there?

As we evaluate our regional peers, we anticipate that the median for the group we consider for compensation purposes at the end of the first quarter was around 10.4. I am not sure we will reach 10.4; I believe we might aim to remain slightly above that. However, we have some flexibility to grow or continue repatriating capital, and our course is largely set through the CCAR cycle. Looking ahead to next year, my long-term expectations for targets likely fall between 10.4 and 11.

And I would remind you, Geoffrey, we've mentioned in the past, yes, we publish our semi-annual, kind of DFAST results and the loss content in our model is a lot less than the loss content in the Fed's model. So in terms of what we think we need to hold, right now we're being constrained, I would say through the Horizontal Capital Review process, but in fact, we think we can hold less capital than that, as Harris said.

And the Horizontal Capital Review process, what exactly do you mean by that?

Oh, sorry, that's the process formerly known as CCAR for banks of our size and complexity. The name has changed; I don't know if the headlines have caught up with it yet.

Okay. And then, but surely on the 10.4, I mean the peer group is going to move down over time, isn't it, as other banks are paying out 100% of earnings and growing loans a bit. So I mean, isn't there, as a peer group moves down, then that's got to get lower than that?

Yes, I had mentioned 10.4, but the actual figure is 10.7, which is a 30 basis points difference. I believe the peer group may decrease, and our approach will reflect that. The key point is that we aim to stay at or slightly above the peer group. There are various ways to analyze this, including the results from our DFAST and our internal models. We want to ensure that in a downturn, we have sufficient capital and do not fall behind, which is something we'll continue to assess. My main takeaway is that we still have some flexibility.

Hey Paul, you've mentioned feeling hopeful about deposit growth going forward. We're just wondering, what you think the shape of that looks like over the next year. And if you think moderate deposit growth is something you can achieve in the current environment, so something in the mid-single-digit range.

I think mid-single digits is going to be a little tough, to be honest with you. That being said, as I've said, we've actually been more successful over, certainly since I've been here, than our business leaders expected. We've got a lot of very high-quality relationships that we continue to attract money into the bank. But the environment is changing, and it's kind of hard for me to be really certain that we could achieve a deposit growth in the mid- to mid-single digit range. I would like to do that; that frankly is probably going to be the high end of what I would expect.

Okay. And then just another question, and then real quick if I could. The loan book yield ex the recoveries was around 4.23. That was up about 8 bps if I back out recoveries this quarter and the reversal last quarter. Can you talk about the accretion of prepayment penalty income trends in 2Q that may have impacted that trend this quarter?

It wasn't a measurable amount. We tried to highlight what we considered the main factor, which was the recovery income. The rest of the increase, as you noted, is mainly due to the change in short-term rates. Referring back to the slide in my comments where we discussed the expected changes in the coupon linked to the change in the index, you'll find that if you do the calculations, it seems to be accurate.

Sure, could you provide some insights on the performance of the new small business platform? You recently introduced it as a small business loan-generating tool, and I noticed there was a good increase in pull-through from that platform. Additionally, can you share how much that contributed to growth this quarter?

I have Michael Morris, our Chief Credit Officer, talk about that. Michael?

Thanks. We call that the business banking loan center, the BBLC. They've had very solid success, both in the approval and pull-through rate. We've added balances of about $11 million in a pretty short time period, and we're talking about...

Roughly $11 million, excuse me, $11 million. Sorry. Anyway, the platform is up and running. It's running well. We like what we're seeing from a credit perspective. We like what we're seeing from a yield perspective. And certainly, the turn time within that platform has been greatly shortened.

Just wanted to follow up on the provision guidance of around $10 million. You guys have had, at least this quarter, a pretty healthy dose of recoveries. Is that also a part of your calculus? I know those are hard to anticipate, but just kind of curious kind of what the pipeline for further recoveries if there was a big driver this quarter, if that kind of plays into some of your guidance there as well?

I'll start up out, and then maybe my partners here can join in. I will remind you that it's not explicitly $10 million, it's sort of $10 million plus, so maybe a little more than $10 million. Yes, there are recoveries in there. But yes, we're looking at the credit quality of the portfolio, and we generally have a pretty decent insight into the sort of upcoming charge-offs, that's generally true. And so our expectation is the credit performance continues to be so strong, that we could modestly adjust that outlook for provision downward from what we've seen in the first quarter.

This is Scott. I would just add that as you noted, projecting recoveries is very episodic and difficult to predict. However, when we examine the levels of energy charge-offs we've experienced over the last 24 months, any standard recovery rate applied to that should lead to a higher level of recoveries than we saw in 2014 or 2015.

I have a follow-up question. Scott, could you provide additional insight on Amegy? It appears that it contributed around 40% to the loan growth this quarter, particularly a significant portion in commercial and industrial loans, excluding oil and gas. I'm interested in the trends you observe there, especially if there are notable segments in commercial and industrial or any other details you could share regarding Texas.

Sure. Happy to do that. One influence was the fact that energy loans were flat as opposed to contracting by anywhere from $100 million to $150 million a quarter, which is what we've been experiencing, so that's the sort of bullet point #1. But more importantly, the math, they've just seen really solid middle market, small business, lower end of corporate growth. And they've really seen that pretty consistently through last year and through the first half of this year. So I would just describe it as good, solid community banking, middle market, private banking loan opportunities, very granular, and of note as well is on the CRE side, there's virtually no new underwriting going on there. There is some, but not, it's not adding to loan balances materially. The 1-4 Family business is doing very well in Texas, and we've not seen any real energy impact on FICO scores, etc., but a little bit of deterioration, but not much. So it's just good, granular community banking type activity.

Around the national commercial real estate portfolio, what is the size of that now? How long before that likely stabilizes?

Yes, hi Jennifer, this is James. We've got $1.8 billion at this time, and maybe I'll turn the call over to Michael here to talk about some changes in policy on that.

Yes, I would add that the national real estate portfolio average balance is below $1 million per ticket. So it's very granular, it's starting to slow down, the attrition is starting to slow down. We've made a couple of changes to the platform, and it should invite some new loan growth, but nothing that would rock the charts. But it's a platform that we will stay with and continue to grow where and when we can.

I would be hopeful that we're within probably 3 to 4 quarters of it really stabilizing and starting to grow. That would be my best guess, but it's purely that. It's largely a function of how much liquidity is out there, particularly with the Community Banks we like. We sourced a fair amount of this product from a network of Community Banks around the country, and they're all very liquid. And so that slowed the originations, but it's getting to a point where it will start to stabilize.

Okay. Follow-up question. You mentioned, Paul, that you're in the very early innings of supply chain cost improvement initiatives. Any way you can quantify the opportunity there?

Yes, Jennifer, and I think you probably heard me talk about this before. Unfortunately, I can't quantify that yet. I will say, we're kind of completely modernizing our supply chain process. And I expect that there's going to be a lot of efficiencies to come out of that. And we highlight that as one example, because we get the question a lot around keys. It seems like you guys have really controlled expenses over the course of the next couple of years. How much longer can you do that? And the point of highlighting that is to say that, there continue to be opportunities as we have consolidated the charters 18 months ago, continuing to work through kind of back office and other operational efficiencies, there continue to be opportunities to continue to make substantial investments in the business while controlling expenses. And so I can't really quantify that 1 without probably going through a whole other laundry list of all other things, including things we might be investing in. So I don't want to mislead you.

Two quick questions. I guess the first in terms of the oil and gas portfolio. You talked about it being stable. Is that due to less runoff of the existing book or is that due to more willingness for you guys to actually put on new loans?

Ken, this is Scott. We have initiated some new underwriting in both midstream and upstream sectors. However, the services portfolio has generally experienced slower payoffs. If we look back to 18 months ago, at the end of 2015, the outstanding energy services loans have decreased by approximately 40%. This decline is mainly due to the fact that around two-thirds of that portfolio consists of term loans which are being paid down over time, and while charge-offs have had some effect, the impact from amortization is significantly greater. Thus, I can say that we are seeing some new underwriting in midstream and upstream, accompanied by an appropriate slowdown in paydowns. Moreover, we are starting to see an increase in revolver usage as the rig count has nearly doubled over the past year.

Got it. Okay. Helpful. I have a question for Paul regarding the tax rate. If I understood correctly, your guidance for the next four quarters is 34% to 35%. How should we consider the impact of the first quarter stock-based accounting change that affects the tax rate in the first quarter? I assume this will occur again in the next first quarter, correct?

Right. Well, it's maybe not necessarily the next four quarters, as you pointed out. The impact is dependent on the strike price of the equity-based compensation when it was issued and the prevailing stock price at that time. What I meant to communicate was that, excluding the effect of that accounting guidance on stock-based compensation, we expect the tax rate to be 34% to 35%. The reason is that it's really difficult for me to predict what that accounting guidance figure will be. Hopefully that makes sense.

On the loan growth and on the commercial side, I'm just trying to get a better feel of how much of that is a pickup in demand, where you're seeing pipelines better and maybe CapEx driving growth and everything versus anything you're changing? Because there's a real notable pickup in growth here, and then I'm just curious if any of that is pricing or any change in your willingness to lend in any way?

If I could quickly jump in, John, this is James. The origination volume we recorded in the second quarter of 2017 was almost the same as in the second quarter of 2016. The main difference was in the previous quarter. In the first quarter of 2017, we experienced approximately $700 million less origination volume compared to prior first quarters, and we don't have a clear explanation for that. However, we and the entire industry faced a similar issue. Therefore, I believe it was more of an anomaly related to the first quarter rather than what we've observed in the second quarter.

And this is Michael. I'll add that demand probably has picked up a little bit. It's probably a combination of demand and process improvement, but not really expanding the underwriting box.

And we certainly haven't changed our pricing perspective as well, either. We're maintaining the same pricing approach as we've had in recent years.

Thank you very much for joining the conference call today to discuss the results. We look forward to seeing you at a conference sometime soon or at the next earnings call out in October. Thank you very much.