Community Financial System Inc (CBU) 2015 Q2 法說會逐字稿

Welcome to the Community Bank System's second-quarter 2015 earnings conference call. Please note that this presentation contains forward-looking statements within the provisions of the Private Securities Litigation Reform Act of 1995 that are based on current expectations, estimates and projections about the industry, markets and economic environment in which the Company operates.

Such statements involve risk and uncertainties that could cause actual results to differ materially from the results discussed in these statements. These risks are detailed in the Company's Annual Report and Form 10-K filed with the Securities and Exchange Commission.

Today's call presenters are Mark Tryniski, President and Chief Executive Officer; and Scott Kingsley, Executive Vice President and Chief Financial Officer. Gentlemen, you may begin.

Thank you, Nikki. Good morning, everyone, and thank you all for joining our call this morning.

Second-quarter results were good and generally in line with our expectations. Last quarter's substantial underperformance in lending was this quarter's strength. Loans were up $100 million for the quarter, led by business lending, with contribution from all markets. Auto lending was strong as well, with mortgage home equity and direct branch lending also up for the quarter, and asset quality continues to be very good.

Fee income results were mixed. Wealth Management and banking were flat, but Benefits Administrations revenues were up 8% over last year's quarter. Expense management was very good, with total operating expenses up less than 1% over 2014. Overall, it was a solid operating quarter.

Last quarter, we discussed the pending Oneida Financial acquisition in detail, so I will just comment on current status. We initially expected the transaction to close this month, but as we announced in June, we've extended the closing, pending completion of regulatory approvals. Oneida shareholders overwhelmingly approved the transaction at their June 17 meeting.

Integration efforts are underway and proceeding well, and we expect to close in the fourth quarter. We are excited about the prospect Oneida Financial brings to our combined organization, particularly around the opportunity related to their Insurance Benefits Administration and Wealth Management businesses to continue to expect 2016 accretion to approximate $0.07 per share on a GAAP basis and $0.11 per share on a cash basis.

Post-Oneida, total assets will approximate $8.5 billion. Given the higher hurdles in many respects of being a $10 billion bank, we have begun to build a runway to prepare for that eventuality. We are making infrastructure investments in operations, capital planning, risk management and compliance in order to be fully prepared to meet higher organizational expectations. The impact of Durbin is amongst the most challenging of these, and we remain committed to hurdling the $10 million threshold in a manner that is not dilutive to our shareholders.

We have also been spending a great deal of time with focus on interest rate sensitivity. As we look back on the most recent increase in Fed funds from 2004 to 2007, our deposit costs rose 8 basis points over the first 100 in Fed fund increases; another 14 basis points over the next 100; and maxed out at 126 basis points of the total rate hike of 350 or 36%.

Over that 11-quarter period, we also saw our total checking and savings accounts decrease very modestly from 48% of total deposits to 44% of total deposits, with time deposits increasing from 41% to 46%. In other words, a 350 basis point increase in Fed funds improved our mix -- excuse me, moved our mix from core to non-core by only 4% to 5%. It's difficult to know if this time will be similar or not, but our core counts have improved from 44% to 64% of total deposits, and we believe we are very well-positioned to benefit from an increasing rate environment.

Lastly, with respect to our outlook for the remainder of the year, we do expect modest core margin contraction with growth in asset generation in banking and non-banking revenues. Asset quality is particularly strong right now, and we expect that will continue through the remainder of the year, and our focus on expense efficiency will continue. We will remain in a significant surplus capital position -- even after the Oneida transaction -- and we will continue our focus on opportunities to deploy that capital in a manner that is productive and additive to shareholder value.

Scott?

Thank you, Mark, and good morning, everyone. As Mark mentioned, the second-quarter of 2015 was a very solid operating quarter for us, with modest yet still productive year-over-year operating improvement trends. I'll first discuss some balance sheet items.

Average earning assets of $6.86 billion for the second quarter were up 3.5% from the second quarter of 2014. Average loans grew $90 million year-over-year or 2.2%. Ending loans were up $99.7 million from the end of the seasonally-challenged first quarter of this year, with productive growth in all of our portfolios.

Average investment securities were up 5.6% compared to the second quarter of 2014, principally a result of our decision to early-invest the expected net liquidity of the pending Oneida Financial transaction. Since early March, we have purchased $400 million of treasury securities, with blended yields just below 2%, as a planned replacement for the $300 million of securities currently held in the Oneida portfolio, as well as the redeployment of the remainder of our own expected investment cash flows for 2015.

Average deposits were up 2.0% from the second quarter of last year. And the multiyear trend away from time deposits and into core checking, savings and money market accounts, continued in the first half of 2015, resulting in a further decline in overall funding costs. Quarter-end loans in our business lending portfolio of $1.3 billion as of June 30 were $56.4 million or 4.5% above the end of the first quarter, and are 2.6% above year-end 2014 levels.

Asset quality results in this portfolio continue to be very favorable, with net charge-offs of under 11 basis points of average loans over the last nine quarters. Our total consumer real estate portfolios of $1.95 billion, comprised of $1.61 billion of consumer mortgages and $341 million of home equity instruments, were also up on a linked quarter basis, and were 1.5% higher than the end of the second quarter of last year, consistent with general market demand.

We continue to retain in portfolio most of our short and mid-duration mortgage production, while selling secondary eligible 30-year instruments. Asset quality results continue to be very favorable in these portfolios, with total net charge-offs over the past nine quarters of just 8 basis points of average loans.

Our consumer indirect portfolio of $837 million was up $33 million from the end of the first quarter of 2015, historically consistent and in line with seasonal demand characteristics. Despite solid new car sales, used car valuations were the largest majority of our lending as concentrated, continue to be stable. Net charge-offs in this portfolio over the past nine quarters were 27 basis points of average loans, a level we consider very productive.

With our continued [rise] toward A&B paper grades and the very competitive market conditions in this asset class, yields have continued to trend lower over the last several quarters. We have continued to report very favorable overall net charge-offs results, with the first half of 2015 results at just 0.06% of total loans being a stellar performance.

Nonperforming loans, comprised of both legacy and acquired loans, ended the first quarter at $23.0 million or 0.54% of total loans. Our reserves for total loan losses represent 1.1% of our legacy loans and 1.06% of total outstandings, and based on the trailing four-quarter results, represent over nine years of annualized net charge-offs.

As of June 30, our investment portfolio stood at $2.87 billion and was comprised of $234 million of US agency and agency-backed mortgage obligations, or 8% of the total. [$663 million] of municipal bonds or 23%, and $1.89 billion of US treasury securities or 66% of the total. The remaining 3% was in corporate debt securities. The portfolio contains net unrealized gains of $57 million as of quarter-end, a level consistent with the end of the second quarter of 2014.

Our capital levels in the second quarter of 2015 continue to be very strong. The Tier 1 leverage ratio stood at 10.20% at quarter-end. And tangible equity to net tangible assets ended June at 8.63%. Tangible book value per share was $15.96 per share at quarter-end, and includes $37.7 million of deferred tax liabilities generated from tax-deductible goodwill, or $0.92 per share.

Shifting now to the income statement, our reported net interest margin for the second quarter was 3.76%, which was down 18 basis points from the second quarter of last year, and 7 basis points lower than the first quarter of 2015. The decision to early-invest the expected liquidity from the Oneida transaction into treasury securities clearly contributed to the overall decline in net interest margin in the second quarter.

Also, consistent with historical results, the second and fourth quarters each year include our semiannual dividend from the Federal Reserve Bank of approximately $0.5 million, which added 3 basis points of net interest margin to second-quarter results compared to the linked quarters. Proactive and disciplined management of deposit funding cost continue to have a positive effect on margin results, but has generally not been able to fully offset declining asset yields.

Second-quarter noninterest income was up modestly from last year's second quarter and seasonally above the first quarter of 2015 as expected. The Company-deployed Benefits Administration and Consulting businesses posted an 8.4% increase in revenues from new customer additions, generally favorable equity market conditions, and additional service offerings.

Our Wealth Management group revenues were essentially even with a very strong second quarter of 2014. Seasonally, our second-quarter revenues from deposit service fees were up 6% from the levels reported in the first quarter, but were up only modestly from the second quarter of 2014, as higher car-related revenues were almost completely offset by lower account overdraft protection utilization. Mortgage banking and other banking services revenues declined $800,000 from the second quarter of last year, which did include nearly $0.5 million of nonrecurring life insurance-related gains.

Quarterly operating expenses of $56.0 million increased $884,000 over the second quarter of 2014, and included $361,000 of acquisition expenses. Excluding those costs, operating expenses were up less than 1% year-over-year, and included merit-based personnel cost increases. Seasonally, as expected, our facilities-related costs in the second quarter were lower than the linked first-quarter, and remained at levels consistent with the second quarter of 2014.

Our effective tax rate in the second quarter of 2015 was 30.5% versus 29.9% in last year's second quarter, a reflection of a lower proportion of tax-exempt income to total income, as well as certain statutory changes driving up our effective state tax rates. We continue to expect net interest margin challenges to persist for the balance of 2015, as many of our existing assets are still being replaced by new assets with modestly lower yields.

As I mentioned earlier, our decision to early-invest the planned liquidity from the pending Oneida Financial transaction certainly contributed to a lower net interest margin for the quarter, but also clearly added incremental net interest income. As such, a small portion of the $0.07 of expected GAAP and $0.11 of expected cash earnings accretion from the Oneida transaction were realized beginning in the second quarter.

Our funding mix and costs are at very favorable levels today, from which we do not expect significant improvement. Our growth in all sources of recurring noninterest revenues has been positive, and we believe we are positioned to continue to expand in all areas. While operating expenses will continue to be managed in a disciplined fashion, we do expect to continue to consistently invest in all of our businesses.

As we frame our expectations for the second half of 2015, we remind ourselves that the third and fourth quarter each year contain one more calendar day and one more payroll day than the second quarter of the year. We continue to expect Federal Reserve Bank semiannual dividends in the second and fourth quarters each year.

Our first-half net charge-off results were extremely positive. And although we do not see signs of asset quality headwinds on the horizon, we would expect higher levels of provisioning for the remainder of the year. Tax rate management will continue to be subject to successful reinvestment of our cash flows into high-quality municipal securities, which have been a challenge at times during this period of sustained low rates. However, we believe we remain very well-positioned from both a capital and operational perspective for the expected Oneida Financial integration in the fourth quarter of this year.

Now I'll turn it back to Nikki to open the line for any questions.

(Operator Instructions). Alex Twerdahl, Sandler O'Neill.

Hey, first, Mark, you're talking in your prepared remarks about beginning to build up expenses in the expense base and building out some of the systems, and adding into various departments as you approach the $10 billion mark. Is that something that's changed? I thought that was something that's been kind of going on for a couple of quarters already at this point. And based on what you were talking about a couple minutes ago, is that going to -- should that change anyone's expense expectations for the coming quarters, and into 2016?

Yes -- no, I didn't intend to signal that it was an increase in expense in the expense base. I think we have begun, during the first quarter, I would say, to invest in resources and software applications and processes, and the consultants and the like, to try to prepare the ground for the $10 billion mark.

So it was just a reminder that that effort will continue as we build out this runway, so that we are fully prepared at the point we do our asset -- the threshold of $10 billion. So it's not intended to signal anything other than as a reminder to our shareholders we are continuing to invest and to think about it, and to be prepared when that eventuality arises.

Okay, great. That's what I thought. And then maybe you can just give us a little bit more color on the commercial lending trends in the second quarter? I know you mentioned that the pipelines were very strong at the end of March. Maybe you can talk a little bit about whether or not the growth in the second quarter was really just a result of some closings being pushed into the second quarter, if there was a change in addition of some initiatives? Or anything along those lines. And then tell us where the pipelines were at the end of June.

Sure. To answer your question, I think that the performance in the second quarter was a result of both of these items and elements that you mentioned. One is there was certainly, if you look at the performance in the first quarter, a lot of things that didn't get done. There were a lot of -- there was deferral of projects, construction and otherwise from the first quarter into the second quarter -- that was part of it.

I think it was also organic activity in the second quarter. So I think the performance and the turnaround, plus $56 million in the second quarter, was partially attributable to deferral and partially attributable to improved activities. I think also I would applaud our commercial business lending team. It was also the result of really very good execution on the ground in all of our markets.

We saw very good activity and growth in our Northern New York markets and our Northeast Pennsylvania markets, and our kind of Central New York markets, and also in our Western markets. So, it was just a really good, strong, all-around performance across the Company in the second quarter.

If you look at where it was, it was really -- it was also spread out across the board. It was CRE and it was also C&I. It was construction; it was some refi's; it was new CRE, some owner-occupied, some not. It was also improved line utilization, particularly by some of our larger commercial customers. So, it was really a -- an exceptionally good and strong performance by the commercial lending team.

The pipeline is -- it remains very strong. I think relative to where it was in March, it's at or above that level right now. So we are expecting another strong performance, let's call it, in the third and fourth quarter. So, the trendline right now in momentum in commercial is very good, and we are hopeful that it continues through the remainder of the year.

Fantastic. That's great commentary. Thank you very much.

Thank you, Alex.

Joe Fenech, Hovde Group.

Guys, on the 1Q call, you indicated that you still considered yourself to be in a surplus capital position post-the United deal, which would seem to imply that you would consider additional acquisitions in the near-term. If that assumption is correct, how does the delay impact your thinking on future deals, if at all? And how does that tie into your comments earlier about preparing for the approach to $10 billion?

Well, I think it's certainly true that we will have considerable less [asset] capital after the Oneida transaction closes, possibly in the $100 million range. So, we will -- as we've suggested in the past, we need to continue to be disciplined to deploy that capital in a way that creates sustainable value for our shareholders. And we will continue to do that.

I don't know that the delay in the Oneida transaction is going to affect significantly how we think about other transactions, other than thinking about the establishment of a closing timeline and the impact of the current regulatory process for reviewing M&A. And I think, historically, we would like to sprint to the finish line and get them announced and closed, execute and move forward.

I think in the regulatory approval environment today, we are going to need to rethink the timeline of that process. And we always used to like to sprint to the finish line; now maybe it's, instead of a 100 meter dash, maybe it's a quarter-mile at this point. So, I think that's probably the only takeaway, Joe.

I think the other element is just, I think, the regulatory environment also, the expectations around things like risk management and compliance, and CRA and fair lending, and BSA, AML and all those things, I think the bar just continues to get raised on banks -- all banks, not just banks that are growing through M&A, but I would say particularly through banks that are growing through M&A because of the opportunity that it creates for regulators to ask more questions.

And I think we need to be mindful of the increased level of questions that might be asked, and be prepared to efficiently answer those questions, as we are going through presently with the Oneida transaction. So, I don't expect a revolution in terms of how we think about integration and the timeline related to that, but I do think it's an evolutionary process where we have to be mindful of what we expect are going to continue to be greater expectations on the part of the regulators as it relates to the regulatory approval process.

Okay. And would you say, guys, there's any hesitation in what I will call tiptoeing up to the $10 billion threshold? In other words, are we more likely to see you try to vault past $10 billion in a significant way? Or would you be comfortable sitting right around $10 billion for a period of time, if that's how the situation turned out to be?

Well, I think if you look post-Oneida, we will be at somewhere in the $8.5 billion level of assets. Ideally, I think, looking at another high-value acquisition opportunity, that would put us in the $9.5 billion range, or maybe slightly larger, and then prepare for a threshold transaction, which might be a little bit larger.

And the closer you are to $10 billion, the smaller the acquisition would need to be in order to hurdle the -- what's, for the most part, the Durbin impact. So, from our perspective strategically -- and we had a lot of dialogue with our Board about this as well -- the idea of getting to $9.5 billion or so, and then thinking about that threshold transaction, is probably the best risk/reward strategy for us that we are thinking about presently.

So, that is the strategy. If we get this $9.5 billion, we can grow organically at 3%, 2% to 4% for a couple years before we would be, essentially from a shareholder value perspective, be forced to think about a threshold transaction. We want to try to kind of strategically prepare how we're going to execute on that.

And we've committed all along that we expect to do so without any impairment of shareholder value, meaning we expect the act of hurtling the $10 billion will create sufficient earnings capacity that it will offset -- at least or more than offset the impact of Durbin. So that's where we're kind of just at right now, Joe.

And then appreciating that there might be sensitivities around what you can say, but can you talk in a little bit any more detail that you can give us on the reasons for the delay?

Well, it's right now with the DC Fed. And they have asked some additional questions around lending and HMDA data, and our -- you know, specific kind of risk management/compliance processes. It was -- the review was precipitated by a public comment letter written to the Fed during the public comment period, who is a social activist from New York City, and has a 20-year history of making the same allegations against other banks as part of the M&A process.

So, the Fed has asked us -- the DC Fed has asked us some questions. We have responded with further information. And we, at this juncture, expect the transaction to close in the fourth quarter.

Okay, great. Thank you, guys.

Thanks, Joe.

William Wallace, Raymond James.

As I kind of look at your decision to go ahead and lever up a little bit ahead of the anticipation that you'll have some excess liquidity related to Oneida, was there anything that you saw in the markets that drove that decision? Or is just really a matter of going ahead and taking that step, you know, you feel, given some certain level of comfort that you have, that the deal will differently close?

Yes, I would say a little bit of both, Wally. I would say that we had -- after we announced the transaction, we had planned for sort of an orderly move to a group of assets that we were more comfortable in our portfolio with, over a longer-term basis, you know, versus where they were positioned today, which is not necessarily a bad position, but it didn't necessarily fit all the criteria that we would be looking for going forward.

So, we did start that initiative in March, continued that into the early parts of the second quarter. And in fairness, didn't really change our positioning on the expectations after we found out the transaction was going to need to be delayed a little bit. So we are still pretty comfortable that when the transaction closes, there would be the disposition of the lion's share of their portfolio; we would just pay off short-term borrowings that, as you acknowledged -- or as you mentioned, we created in the second quarter.

I think there is a little bit of market timing with us, relative to certain indicators that we saw in the market that said, this group of assets, based on their duration and risk characteristics, is essentially a productive time for us to be a buyer. So we exercised on those timings. But again, I would say, more than anything, we have sort of put on $300-million-plus of the expected $850 million of assets we will get.

So, to Mark's point, you shouldn't expect to see the balance sheet grow $850 million by the time the end of October comes around; probably more like $550 million.

Okay. Perfect. And then the borrowings that you guys put on to fund the strategy, will you keep your borrowings and get rid of some of theirs? Or will you just use theirs and get rid of whatever you put on?

Yes. You know, we might. They have very little borrowings themselves. Matter of fact, they're actually in a net cash position, as we speak. So, the expectation is we would extinguish our short-term borrowings in the current mindset.

Wally, that being said, we will have plenty of capital to sort of stickhandle through that after the transaction. So if it does look like it's productive to carry a little bit of leverage into 2016, we are certainly not afraid of that either.

Okay. Okay. And then just switching gears a little bit. The decline on a sequential basis in your mortgage banking fee income line was a little bit surprising, just given the seasonality we typically see in the second and third quarter, and what we are seeing in other banks. Is there anything going on there that would've driven your production lower? Did you balance sheet more or --?

Yes. But let me run through that a little bit because I think that's probably the fourth question this morning I've gotten on that. So, I think if I start with the mortgage banking in the second quarter of 2014, essentially from mortgage banking-related activities in the second quarter of 2014, we were about $350,000 of revenues compared to $200,000 in the second quarter of 2015.

Essentially, we took $130,000 mortgage service impairment charge in the second quarter of this year. It doesn't sound like the accounting in the rate environment would ever lead to that outcome, but as you know, that's a very mechanical, calculated outcome. So essentially, some of the production we sold in the last two years reached its season capability or reached its seasoned destination.

And rates are modestly higher today, so there is a sense that there is still some modest prepayment risk attached to that. And we carry a very, very small asset of associated mortgage servicing rights. So that's a piece of it.

To your question relative to the third and fourth quarter of last year that had higher activity there, a couple other things to point out. We typically are a larger seller of mortgage transaction in the second half of the year every year, just based on activity. In other words, fourth-quarter pipeline isn't terribly robust for first-quarter activity. And we talked about first-quarter activity being very slow this year, so second-quarter sales were pretty small. We would expect a higher number.

The other piece that ends up on that line from a presentation standpoint is, we end up with sort of other banking services revenues, which, for us, tend to be BOLI-related income, small miscellaneous stuff that can't find a home elsewhere. But the last year in the second quarter, we had about $450,000 of nonrecurring life insurance gains coming out of some of our BOLI assets, and certainly don't expect or project those to continue.

But on a going forward basis, I think if you used the number that we have in the second quarter, put back the mortgage servicing impairment charge that we wouldn't expect to be recurring, you're kind of running at a level that's in that $900,000 to $1 million per quarter. As a reminder for everybody, we do expect a dividend from one of the retail insurance programs we participate in on a [pooled] basis that always come from the third quarter.

We always say it's in and around $0.01 per share. And that number could move up or down $100,000 or $200,000 in any given year. But that should be an expectation for people's modeling for the third quarter.

Okay. And thanks. And Scott, just one kind of housekeeping question. In your prepared remarks, you mentioned, as it relates to margin, that this quarter had a 3 basis point benefit related to something that was not expected. What was that? I missed it.

Yes. So the Federal Reserve, Wally, pays you your semiannual dividend on your holdings of Federal Reserve stock -- June 30th and December 30th each year. So that $500,000 for us or $480,000 for us shows up in the second quarter and the fourth quarter. So you tend to get, quote, a kick-down in the third quarter on yield just because that's out there.

I appreciate it. Thanks, guys.

Very good. Thank you.

Thanks.

David Darst, Guggenheim Securities.

Just -- Mark, that's really helpful on the liability duration and deposit pricing details you gave us. But how are you thinking about the asset side and your AOCI risk kind of rising in your environment. And were there -- are there not other alternatives relative to the treasuries you bought that would have maybe a better profile for rising rates?

You know, David, I think that you could sort of have to preface that by saying we are certainly not managing the institution with fluctuations in AOCI. As you know, from a regulatory standpoint, any bank under $250 billion in size opted out of including changes in AOCI as part of their regulatory capital, as it was. So not really something we are thinking about from a day-to-day operations standpoint.

So the only time you get any kind of discussion around AOCI changes in a rising rate environment would be similar to what you talk about in interest rate sensitivity. So in other words, if something could happen, are you going to put yourself into an unrealized gain position or build an unrealized gain position to an unrealized loss position with a rate change. And we understand, given the size of our portfolio, we are certainly susceptible to some movements there.

But since that's tangible equity that you really can't use from an operating standpoint, it is not something we spend a lot of incremental time thinking about. It's out there for purposes of enterprise value calculations from the EVA side of stuff from a regulatory standpoint, but really we don't get a lot of questions on it, David.

I think what you're kind of back to in your discussion or your question, though, would be saying -- did you have a choice of another asset class that had different duration characteristics than the treasuries you pulled? We picked treasuries that were a little bit above sort of a five-year duration. And we thought that those fit into our longer-term composite balance sheet type of a planning attribute -- not lost on us with they are qualitatively superior to almost every other choice we had out there in the market today.

And similar to Mark's comments, I think we looked at what we thought would be a slow gradual pace of interest rate changes on the way up. And we looked at our historical and expected performance on deposit retention and deposit pricing characteristics, and thought -- these don't run afoul of any of those kind of expectations. So, on a composite basis, David, that's kind of the justification.

Okay. So should this change our kind of -- you said obviously the EPS accretion doesn't change. But should this change our kind of margin expectation on a pro forma basis for --?

Yes -- I would look at it this way, David. A decent question. I would say that we did bring forward a small portion of the earnings accretion from the transaction. I mean, any time you can put incremental $300 million of assets on the balance sheet, with the yield differential between your 200 basis points of yield on the instrument and 45 basis points of borrowing costs, you've picked up some income on an early basis.

On the margin side, I would say we had been already signaling that Oneida carried a lower net interest margin base than we did pro forma, so there would be some dilution to that interest margin in the combination anyways. Kind of think of it this way now -- bring on now $550 million of a balance sheet for Oneida, that essentially does not have investment securities on it today.

So, that's where I would kind of go with my pro forma side, that essentially we are bringing over a loan portfolio that has similar characteristics to our kind of yield on the loan side. Their deposit pricing is a little bit higher than ours today, because in fairness, at their size, they don't enjoy some the interest rate sensitivity flexibility that we do in their primary markets.

But if you're going to get a mark-to-market of all that stuff as of the closing day anyways, so I would essentially say, yes, think about getting a $500 million balance sheet that's probably got net interest margin characteristics in the [335] to [350] level. Mix that with what you just saw from us at the current [376], you'll probably still get a little bit of margin dilution.

Okay, got it. And then just on credit in your provision outlook, I mean, should we be -- should we really kind of think about you being kind of sub-10 basis points for next couple of quarters?

Well, I think we would love to book it for the rest of our careers, David, but from a capturing standpoint, I kind of look at it this way -- that we essentially have been providing for a provision about net charge-off levels very historically. And we have enjoyed usually serving most second, third and fourth quarters, enough organic growth to validate why you continue to book above the charge-off levels.

All the indicators are very, very positive today. So if you ask me if I thought I need to retain at a 106% going forward, that number could be 105%, could be 104%. But again, it will depend on the underlying performance of the actual portfolio.

In our second quarter, we enjoyed some pretty good rating changes on the commercial side, which again, led to a conclusion that you just didn't need as much as you did before, from an overall proportional outcome. But boy, if we could sign up for 3 or 6 or 9 basis points of charge-offs and validate that for the next three to six quarters, we would love to do that.

Given the mix of our portfolio, it would be hard to project that. But at the same point in time, nothing on the horizon that we see that's leading us to believe we should hunker down for anything significantly higher.

Okay, great. Thank you.

Thank you.

Collyn Gilbert, KPW.

Mark, just back to your commentary on the crossing of the $10 billion. As you guys kind of look at your budget process, is there a timeline that you think you would achieve this $10 billion crossover? I know you'd said deal $9.5 billion, grow organically 2% to 4%, and then maybe something more meaningful post that. But just trying to get an overall time-frame as to how you're thinking about this.

Well, I think at this juncture, it's probably several years out, but we think we need to prepare for that, because it's going to take a little bit of time to put in place the improvements or enhancements to some of the risk management processes to get our capital planning in [DFAS] disciplines up to speed. So we are not in a hurry to do it.

We would prefer not to have to do it, but you have to grow if you want to continue to create above-average returns for your shareholders, which means we need to exceed the $10 billion threshold at some point. And we are also faced with dealing with the realities of what that means. So we are trying to do that in advance.

So we are not in a hurry to get there. I don't think that we care if it takes five years. I don't know that we care if it happens over the course of the next three years -- or two years if the right opportunity arises. But the one thing that I will say is that we will be disciplined about how we go about that, and expect again to execute on our commitment to shareholders that we do it in a way that does not dilute shareholder value.

Okay. Okay, that's helpful. And then, Scott, just back to the comment you made on mortgage. You had said that you're retaining mostly the short and mid duration but then selling 30-year. How is that breaking out? Number one, what is the short and mid duration that you are retaining? And then number two, just what is the split that you are seeing between what you are selling versus what you're retaining?

Today, we are probably retaining 75% of our originations. Remember the little bit of a different perspective in our marketplace -- 15-year fixed is the product of choice. It has arguably been the product of choice in our marketplace for probably the last 9 quarters. So I think we have done a lot of work around expected duration of our mortgage instrument.

And interestingly enough, in our marketplace, whether you wrote the mortgage at 15, 20, 25 or 30, and you portfolio'ed it, its expected life is in and around 9.5 years. So what does that actually tell you about the consumer? The 15-year -- the people who select 15-year products are probably trying to get to the finish line in terms of not having a mortgage obligation.

It probably tells you that the people who originally selected 30 are certainly moving either in your market or out of your market in a shorter period of time. In other words, none of these intrusions are making it through that duration.

So I think it leads us to be comfortable that we are not putting on by portfolio'ing 15 and sometimes 20-year instruments, we are not putting on above-average interest rate risk. Because again, the consumer has historically portrayed to us that, indeed, they are not going to go the duration on the instrument.

And, Collyn, we didn't just use the last sort of four, five or six years as the cycle outcome. We actually went back 30. And shockingly, we actually found the data -- shockingly for me. But at the same point in time, that's something that, through multiple cycles, the indicator was 9.5 years should be what you expect out of the duration of the portfolio.

I also think too, we have a smaller average mortgage size that almost every bank in the country -- certainly in our size. So when we start to prepare ourselves for what does it take to build up a group of assets that we are actually going to sell in the secondary market, it usually takes us two or three times the amount of loans. So from a processing and a preparation to closing standpoint, we do have to acknowledge that we better be really efficient at it, because they're small-valued outcomes.

So, I think you've seen with the refinancing side of the world sort of complete, maybe complete in the last year, new purchase money is certainly the majority of the new instruments that we are seeing. I won't say we're not still seeing a little bit of refi activity out there. But generally, you're starting to see people who, during the refi, they are trying to shorten the duration of their remaining instruments with the finish line in sight.

Okay. Okay, that's helpful. That's all I had. Thanks, guys.

Thank you.

Thanks, Collyn.

(Operator Instructions). Matt Schultheis, Boenning.

Really quickly -- most my questions have been answered, but really quickly, could you remind us of the anticipated merger-related costs in the third and fourth quarter?

I think we are using a number in the neighborhood of $12 million, some of which will be borne by Oneida and some of which will be borne by us. In terms of the combination of severance activities, termination of IT-related outcomes, change of control type agreements. So, we haven't really changed from that. Some of the delay in the timing of the transaction may actually result in some of those costs being sort of pushed onto the Oneida financials as opposed to ours. But we really haven't changed our thoughts on that.

Okay. So, you are not willing to share with us what the impact on your income statement would be for the next few quarters? Until it's been resolved, I mean?

We haven't changed our estimate of that. You can use [12] if you want to. It's a nonrecurring expense. I'm fine with that.

All right. Thanks.

Thank you.

And there are no further questions at this time. Gentlemen, I will turn the conference back over to you for any additional or closing remarks.

Great. Thank you, Nikki. I appreciate everyone joining in on the call this morning. And we hope to see you again on our third-quarter call. Thank you.

Thank you. And that does conclude today's conference. Thank you for your participation.