Jan 28, 2013
Executives
James R. Abbott - Senior Vice President of Investor Relations & External Communications Harris H.
Simmons - Chairman, Chief Executive Officer, President, Member of Executive Committee and Chairman of Zions First National Bank Doyle L. Arnold - Chief Financial Officer and Vice Chairman W.
David Hemingway - Chief Investment Officer and Executive Vice President of Capital Markets & Investments
Analysts
Ryan M. Nash - Goldman Sachs Group Inc., Research Division Paul J.
Miller - FBR Capital Markets & Co., Research Division Steven A. Alexopoulos - JP Morgan Chase & Co, Research Division Ken A.
Zerbe - Morgan Stanley, Research Division Brian Foran Joe Morford - RBC Capital Markets, LLC, Research Division Kenneth M. Usdin - Jefferies & Company, Inc., Research Division
Operator
Good day, ladies and gentlemen, and thank you for standing by, and welcome to the Zions Bancorporation Fourth Quarter 2012 Earnings Conference Call. [Operator Instructions] As a reminder, this conference may be recorded.
It's now my pleasure to turn the time over to James Abbott. Please go ahead, sir.
James R. Abbott
Thank you, Ian. Good evening.
We welcome you to this conference call to discuss our fourth quarter 2012 earnings. Our primary participants today will be Harris Simmons, Chairman and Chief Executive Officer; and Doyle Arnold, Vice Chairman and Chief Financial Officer.
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 dealing with forward-looking information which applies equally to statements made in this call.
A copy of the earnings release is available at zionsbancorporation.com. We intend to limit the length of this call to 1 hour, which will include time for you to ask questions.
During the Q&A section, we ask you to limit your questions to one primary and one related follow-up question to enable other participants to ask questions. I will now turn the time over to Harris Simmons.
Harris?
Harris H. Simmons
Thanks very much, James, and welcome to all of you to the call today. Before I start into this, I'm going to go off the script for just a moment and briefly recognize James Abbott here.
He's been recognized by all of you, both buy and sell side, as the best IR exec in the mid-cap bank space by Institutional Investor Magazine, and we're really proud of James and the role he plays here. So we may have -- or had a lot of moving parts in our net interest margin and other things, but apparently, at least, we give James credit for explaining it well, and I just want to congratulate James in front of his constituents here.
Expectedly, the quarter -- several underlying trends of the fourth quarter were generally quite strong, with better-than-expected performance from credit quality, from certain of our capital levels, loan growth, net interest income. We're encouraged with our ability to cut some costs as we move through 2012, primarily from the cost of debt and preferred equity.
We're hopeful that we'll be able to continue to improve our return on equity in 2013 as we intend to further reduce the cost of our capital and financing structure. Let me touch upon a couple of fundamental highlights for the quarter.
Credit quality. First, an 8% national unemployment rate and 2% GDP growth isn't what any of us would call a robust environment, but our credit quality metrics are rapidly returning to levels last seen when the economy was in much stronger shape than it is today.
We attribute a portion of the improvement to improved risk management and credit risk reduction efforts for the past few years and, some of it also, the customers' continued ability to repair their personal and business balance sheets despite the tepid economic environment that we're in. Our net charge-off ratio was only 0.2%, about 20 basis points, a level comparable to the full year of 2007.
Our nonperforming asset ratio, expressed as a percentage of loans and other real estate owned dropped just below 2%, and we expect to see further general improvement in that ratio as we continue through the year 2013. Loan growth.
We generally see more variability probably at loan growth in the fourth quarter than other quarters, and this wasn't any exception. As many of you know, we raised the cautionary flag in late November at an investor conference and as loan balances had declined more than $150 million from the end of the third quarter.
We subsequently experienced particularly robust growth in December, some of which was related to tax strategies on the part of customers. That would have been obviously unique to the fourth quarter of 2012.
Some of it was short-term draws by some of our customers who subsequently paid off a portion of these borrowings in the early part of the first quarter. Excluding the variability that goes with the fourth quarter, our lenders report to us that they remain positive about growth prospects in the loan portfolio.
The pipelines are generally still pretty strong. Customers seem to be generally more optimistic than they were 6 months ago, and production volume rose in each quarter of 2012.
Although pricing remains competitive, we have not experienced much additional pressure on the loan production in the last 6 months. Net interest income declined about $8 million or 2% compared to the prior quarter.
In November, we said that we anticipated perhaps a 2% to 3% decline in core net interest income. And while we haven't calculated that for you in this release, I'll note the decline was less than anticipated.
We currently believe that net interest income should be relatively stable in 2013. And then a final comment, just about convexity risk.
We're seeing increasing signs of economic activity around the country. And while it's unclear of when interest rates will rise, we're one of the very limited number of banks that are significantly asset-sensitive, as many of you know.
Not only do we expect earnings to increase when rates rise, we expect to be able to avoid a significant haircut to equity that many banks may experience due to the heavy concentrations of mortgage-backed securities and other fixed-rate obligations with convexity or optionality in them that we could see in a rising-rate environment. With that overview, I'll ask our Vice Chairman and Chief Financial Officer, Doyle Arnold, to review the quarterly financial performance.
Doyle?
Doyle L. Arnold
Thanks, Harris, and good afternoon, everyone. Let me add my congratulations to James.
And let me also note, I'm suffering from a bit of a cold, so if I happen to go into a coughing spell, I may lateral to James at any time to get through my notes and maybe lateral back at some point. So I apologize in advance if I have to do that.
As noted in the release, we did post net income applicable to common shareholders of $35.6 million or $0.19 per diluted common share for the quarter. As noted in the release and as we had previously disclosed back in maybe the early part of December, we recognized -- we did recognize a significant impairment charge against the CDO portfolio this quarter, as well as gains on securities that also came from CDOs, which netted to an after-tax cost of about $0.25 per share.
Let me also point -- and we'll talk a lot more about that in a few minutes. Let me also point out some -- a couple of accounting and disclosure changes that we made this quarter.
Although the first with regard to some reclassifications, as described in the press release in some detail, I'll note that we reclassified a couple of items which had the net effect of reducing net interest income and increasing noninterest income. We presented all prior quarters on an as-adjusted basis for comparability.
For example, the net interest income is reported at $430 million in the fourth quarter compared to a revised $438 million in the third quarter, whereas we had previously, last quarter, said $444 million. So the net impact of the reclass was $6 million in the third quarter.
The other line item that changed was other service charges, commissions and fees, which increased as a result of the changes. The bottom line, net earnings, pretax or net earnings to common were not impacted at all.
We believe these changes bring us into greater consistency with what is common disclosure practices around the industry today. Also, as Harris didn't highlight but sort of alluded to earlier, this quarter we haven't calculated core net interest income and core net interest margin for you as the primary reasons for calculating that information was just try to strip out some of the noise related to the rather volatile subordinated debt conversions and also some of the income related to loans acquired with FDIC assistance.
Conversion activity has slowed to, recently in this quarter, a minimal level. And the FDIC-assisted loan balances have declined to roughly $0.5 billion from more than $2 billion if you go back to late 2009 when they were acquired.
So we think the adjustment is becoming less meaningful and informative, so we haven't really highlighted that this quarter. But we have given you, I think, what you need, and those who want to can make that calculation.
Okay, turning to revenue drivers. Average loans increased $100 million compared to the prior quarter.
We, of course, pay more attention to average loans because of the variability of -- in the period balances and because of a number that actually drives interest income and earnings for the quarter. In the period, loans increased $463 million, excluding changes in FDIC-supported loans.
But approximately $100 million of that has run off thus far in January. I would note, and we'll talk about it again, that we saw a similar phenomenon that was even more pronounced a year ago of runup and then rundown.
The runup wasn't quite as strong this quarter and the rundown at this point hasn't been as strong either. Let me draw your attention to the loan table on Page 10 of the release.
Commercial and industrial loans had some sub-seasonal trends, while they grew nearly 4% from the prior quarter and some of that backed off in January. I'll point out that C&I loans increased 8% compared to the year-ago period.
And so the general trend remains positive. As we've discussed in the past, the decline in owner-occupied loans is largely attributable to a decision made over a year ago as a part of our concentration risk management efforts to selectively reduce certain aspects of our exposure in international real estate business, the bulk of which is the SBA 504 loan product.
We expect additional attrition in this portfolio throughout this year before the balance is again destabilized late this year or early next. Construction and development loans were down $17 million or approximately 1% sequentially.
Compared to the prior year, the balance is down 14%. We do expect this category to increase in 2013 due to the fairly strong growth in new commitments made in the second half of 2012.
And in fact, balances are up moderately in January. Current CRE declined at 5 of the banks during the quarter, leading to a decline in the overall balance of about $77 million.
This was primarily driven by elevated prepayments, new production volume actually exceeded the trailing 4-quarter average. We do expect this category to grow over time.
But in the very near term, it may decline slightly. One of the headwinds to growing this portfolio is the reemergence of a CMBS market where originations were nearly twice as strong in 2012 compared to 2011.
Finally, within consumer lending, the residential first mortgage loans grew at about 4% linked quarter, which is a rate that is comparable to the prior quarter, and the growth was fairly widespread across footprint. That's kind of a sort of a horizon on -- or the landscape on balances.
Turning briefly to commitments. The unused commitments continue to grow at a solid pace even though the average loan growth of balances was not particularly strong.
Commitments are growing at -- increasing at an 11% compounded annual growth rate during the last 2 years. A meaningful amount of that increase comes from loans, for example, commercial construction loans that are scheduled to "fund up" over the course of the next several months and quarters and thus we believe this should translate into loan balances increasing moderately over the year.
Okay, turning to the net interest margin on Page 14 of the release. You'll note that the NIM declined 11 basis points compared to the third quarter.
The components of that decline are roughly as follows: about 4 basis points were simply due to the increase in average cash balances, which you will note were up sharply despite having wired $700 million of cash back to the treasury right at the end of September; another 2 basis points was due to the decline in yield on FDIC-supported loans; and then the decline in other loan yields adversely impacted NIM by about 5 basis points. As we highlighted a year ago, last January, we didn't think 2012 was going to be a really great year for NIM, and it wasn't.
We noted that it would take a significant amount of loan growth, about $400 million to $500 million per quarter, to offset the NIM effect on net interest income more than -- which was much more than roughly $100 million per quarter of average loan growth that did occur. But we also said that as we enter into 2013, the pressure should subside to a more moderate level.
And we do still believe that's the case. Assuming a static balance sheet, we estimate the NIM should drift down slightly.
However, because a fair amount of resetting and refinancing volume is behind us and because pricing on new loans seems to have somewhat stabilized, the pressure on net interest income over the next 12 months should be much lower than it was in 2012. Coming briefly to noninterest income, just a basic comment there.
There are 2 things worth highlighting, both are related to the CDO portfolio. The first is to the impairment due to higher default probabilities on certain loans and the second is impairment due to higher prepayment assumptions that are most likely to happen in future quarters.
First, let me comment on the higher default probabilities. And that was the biggest component of the most notable and noisy item in the quarter, which was the impairment charge of $84 million on CDOs.
This was discussed at some length in the earnings release, as well as in the pre-releases and discussions by us at the FBR and Goldman Sachs conferences in late November, early December. But let me quickly recap here.
I'll note that this discussion is not going to be an all-inclusive discussion of every nuance of the CDOs. We've tried to lay out that in great detail in our 10-Qs, and we'll be updating it in the 10-K which will be filed in a month or so.
Now it's well-known to you -- most of you that we do own a portfolio of bank trust-preferred CDOs. They're about 800 underlying bank holding companies that are the sources of collateral and cash flow to the trust.
If a bank holding company decides to defer interest payments either because its management elects to do so or because its regulators mandated or arm twisted, however they get there, the bank holding company generally, under terms of the offerings, have 5 years in which they can defer payments without triggering an event of default. If they stopped -- if they continue to defer after 5 years, then it is an event of default.
But if they come current before that, it's not. Each bank holding company has one or more subsidiary banks.
And historically, as the strength of the subsidiary bank goes, so goes the performance of the cash flows to the trust or the CDO. For example, a well-capitalized profitable bank is generally able to make its payments to its parent bank holding company which, in turn, is unlikely to default.
Many of the deferring banks to which we have exposure, about 2/3 of them, are well-capitalized and profitable. However, we recently noted that some small bank holding companies are beginning to use what is known as a 363 Bankruptcy which could render invalid the assumption that a well capital bank has a low probability of default.
Said differently, the subsidiary of the deferring bank holding company might be in decent shape but because the cash may be trapped to subsidiary due to regulatory constraints, the holding company simply may be unable to bring the trust-preferred obligation current at the end of the 5-year deferral period. It's also possible that even if there is cash at the parent company, it may be blocked from flowing out from the bank holding company to the trust by, for example, regulatory orders or things of that nature.
In either case, we, in conjunction with our regulators and consultation with them, recognized therefore that some of the default probabilities on our banks really may be different from the default probabilities of the parent bank holding companies that issued the trust-preferred and, therefore, that our expected loss in some of these cases might need to increase as a result of this new development. However, because the trust-preferred securities are typically the only security issued by the smaller bank holding companies, we generally rank at or near the highest in the liquidation preference, which means we're still entitled to share in the proceeds of a bankruptcy sale.
But if there's a bankruptcy process, then the outcome is less certain. Again, 2/3 of deferring bank holding companies are supported by bank subsidiaries that today are well-capitalized and profitable, which may be called -- result in some recovery value even if they do go through this process or get to the end of the 5-year period.
However, the sale price of the primary asset, stock of the subsidiary bank, may or may not be sufficient to pay off all of the bank's trust-preferreds. There's just many a slip between the cup and the lip, as they say, in how these things may get resolved.
We intend to be extremely active and have been in pursuing our interest in these cases. And although the number of cases we have pursued is small, we and the other significant holders of these notes or trust-preferred securities have generally been successful at recovering much or all of what our model estimates to be the value of the asset.
However, because there's no historical data for re-performance rates and the outcomes of these kind of restructuring and bankruptcy proceedings, other than what we're seeing from our own CDOs, it's also possible we could have to recognize additional OTTI's as new evidence becomes available. Generally, I'd say as a result of all of this, we went back and looked at the banks in our deferring pool, remodeled them and, as a general matter, increased the probabilities of default and therefore the expected loss on, not all, but a number of them, which led to the majority of the OTTI that we took this quarter.
The remainder largely came from impairment due to higher prepayment assumptions because we've been talking about things that might not pay off timely. Also during the fourth quarter, we saw people paying off more than timely.
We saw a significant increase in prepayments by healthier institutions. And therefore the company made a decision to increase the assumed prepayment speeds on performing small banks and here small -- or bank holding companies, and here small is defined as those with less than $15 billion of assets.
Historically, we've observed this CPR speed to be around 3%, which is what we've been modeling. However, the mission we've observed has significant increase in prepayments in the fourth quarter and there are strong arguments to expect continued higher rate for several years as new bank regulations regarding the capital treatment of trust preferred securities are phased in.
Therefore, we increased the CPR assumption in our models from 3% to 10% for the next several years. Because cash flows are generally diverted to senior tranches first, this modeling change has the effect of more quickly paying off the senior tranches in full but results in reduced cash flows from the junior tranches as prepayment reduces excess -- let's call it excess spread, really just future cash available to those junior tranches.
This could result in weaker credit values of the junior tranches, and did so this year. Then finally, we also recognized gains from full payoff of previously impaired securities this quarter.
We recognize gains when either we get a full payoff on previously impaired securities or we get a paydown on any of the securities that we had previously purchased at fair value out of our former QSPE Lockhart Funding. And the gains this quarter were due to the latter.
These were securities that we bought out of the QSPE at a discount to par or amortized cost because that was their fair value, and we subsequently received full value for them. They paid off in cash.
And then finally related to the CDOs, we saw an improvement in the AOCI mark accumulated other comprehensive in part -- income of nearly $90 million after-tax. Some of that was a result of the AOCI mark going to the income statement because of the stuff we just discussed.
And in others, it was just a result of -- resulted from improvement in risk spreads for riskier assets that we observed in the market. For the year, AOCI improved approximately $145 million due to, again, CDO valuation improvement.
Turning to credit. The very short version is that we continue to see strong improvement in nonperforming assets and net charge-offs and other credit metrics.
As a result, our allowance for credit losses model indicated a more moderate -- a moderate negative provision. We did highlight in the release that the primary reason for the sharp decline in net charge-offs was a very strong quarter for recoveries.
Recoveries are expected to continue, but this quarterly weight may have been a particularly high one, and in any event, this is kind of lumpy. Capital, and let's see, that's the last item before we turn to guidance.
The tangible GAAP common equity ratio declined to 7.1% from 7.2% in the prior quarter. And this was due essentially entirely to the significant buildup in deposit balances which were invested in deposits at the Fed, Fed Funds Repos.
The estimated common equity Tier 1 ratio on a Basel I basis declined by about 8 basis points. Deposit balances have declined significantly since year end, which has resulted in a reduction of total tangible assets.
These declines do not appear to be related to the expiration of TAG, and we kind of welcome them in a sense and the reduction should remove pressure on the GAAP capital ratios in the first quarter. And that's, I believe, that's for capital.
Okay, a little bit of guidance and for the next few quarters, kind of how does 2013 look, and then we will turn to your questions. First, loan growth.
With continued strength in the loan pipelines and the increase in commitment in the last 6 months and our customers seemingly feeling a bit more optimistic, we expect continued moderate loan growth over the coming 1 year or coming year. Net interest income.
We do expect the net interest income to be relatively stable, at least in the next few months and quarters. The margin can be volatile due to the cash balances which, as noted previously, declined so far in Q1 after increasing dramatically in Q4.
But on a static balance sheet and a stable rate environment, we would expect some additional compression in the NIM, but offsetting that would be the loan growth that appears is likely. Noninterest income.
We expect the less volatile components of noninterest income such as service fees to continue a modest upward trend. We don't have much income from Mortgage Banking but even as the current refi booms, besides we expect some degree of offset to the slowing refis due to the fact that we're small in the market and we have been increasing our efforts to expand our mortgage lending business.
Noninterest expense. The primary challenges to noninterest expense from the first quarter will be the increase in salary and benefits and the usual tax, social security-related stuff.
Also in the fourth quarter, we reversed about $3 million of previously accrued incentive compensation, as certain long-term plans were no longer expected to payout at the previously accrued rate because of the performance in 2012. So 2012's expense was depressed because of that and we won't have that in the first quarter of 2013.
The fourth quarter of 2012, that is, was held down because of that. So provision expense.
We expect the provision expense to remain low. Continued reduction in problem credits and the ongoing improvement in loss severity rates continue to have the potential to result in a negative provision in the third quarter as it was in the fourth quarter.
Although if loan growth strengthens enough, that could offset the ability to have negative provisions. Preferred stock dividends.
We also, for those of you who haven't seen it, released after market today a press release announcing our intent to issue a new Series G preferred stock, and we expect to be issuing some additional preferred stock during the first half of 2013. We've previously discussed with you the fact that our Series C preferred stock, 9.5%, is callable in the third quarter of 2013 and we're -- and that we would be unlikely to call it without issuing probably not 100% of it in replacement preferred but a significant portion.
So we'll be issuing some of that preferred which will temporarily drive dividend expense up. But in the long term, it should reduce the -- further reduce preferred stock dividend costs significantly.
I believe that's enough of a monologue for me. And we'll -- operator or moderator, if you could queue up the questions, we will endeavor to respond to them.
Operator
[Operator Instructions] And it looks like our first question comes from the line of Ryan Nash with Goldman Sachs.
Ryan M. Nash - Goldman Sachs Group Inc., Research Division
Can you give us some greater details on the outlook for the relatively stable NII? Is that relative to 4Q's level or relative to the entire 2012?
And does the outlook include any of the capital actions that you laid out that will reduce interest expense?
Doyle L. Arnold
I think the comment was relative to kind of 4Q as a starting point, not the whole year. But kind of the basic thing I tried to convey was whereas a year ago we thought net interest income would be trending down pretty steadily throughout 2012.
We think that 2013 will first see a slowing then a flattening and then an improvement as we go through '13. That's what I tried to convey.
Now, in general, there may be some noise in the net interest income and noninterest expense as a result of some of the things that we may do during the course of the year. So those were -- kind of my comments were back to the concept of a core.
And I'll -- maybe without going into get too great a detail and tipping our hand, I mean, to the extent that we issue new debt in replacement of debt to be maturing or what have you, that will temporarily drive interest expense up. To the extent, however, that we tender for debt ahead of its maturity, that temporarily drives interest or other expense up but then drops it significantly.
So what we will do, I want to get away from us computing a core, what we will try to do each quarter in which some of these things occur, if they occur, is give you a quantification of the components so you can make the adjustment that you deem to be appropriate. James, do you have any other comment on that?
Okay, go ahead.
Ryan M. Nash - Goldman Sachs Group Inc., Research Division
Just as a follow-up. I know you talked about the rate of decline and the NIM slowing.
And last year, you had talked about $400 million to $500 million of average loan growth needed to offset. How has that changed as we look into 2013?
So what type of loan growth do we need to see in order to keep NII flat?
Doyle L. Arnold
Significantly less. I don't know that we have an exact number that we've calculated yet.
But do we, James? Do you have a good idea?
But I'll tell you...
James R. Abbott
Probably around 2% to 3%. 2% to 3% loan growth would take care of the margin compression.
Doyle L. Arnold
And the reason is that we have gotten through the largest single period of 5-year resets that -- of loans that were made back in 2007 when rates and spreads -- the rates were much higher. They've all reset.
All of those have reset. By the first half of 2008, originations of those kinds of loans had already begun to trend off.
And by the second half, they've really collapsed as the economy -- well, you don't remember what happened in 2008. So the number -- the amount of loans that are resetting from higher rates that were prevalent 5 years ago to the very low rate environment today will be smaller in the first half, still significant but smaller in the first half of '13 than it was throughout last year.
And by the second half of the year, it will -- that phenomena will be 80% behind us. The other thing is that loan pricing, as I mentioned, has stabilized whereas it's going back to the beginning of last year, the pricing of new loans was still trending down a bit.
It does seem to have stabilized through most of the year. So there are particularly -- there are very specific reasons that we can point to, to say.
And I guess to further it, I think the volume of out of the money options that matured in '12 was also higher than in '13 and '14. There's still some, but all of those headwinds are smaller this year and going forward than they were last year.
Operator
Our next question comes from Paul Miller with FBR.
Paul J. Miller - FBR Capital Markets & Co., Research Division
On your NII, stable NII, what type of interest rate scenario do you have or interest rates don't really matter now because of all the resets running through?
Doyle L. Arnold
I was -- my comments were not predicated on much of a change in the interest rate environment. We remain -- I didn't comment on this, but we remain very highly asset-sensitive so that if the Fed does tighten or, if for any reason, rates were to rise earlier than the Fed has indicated, we believe that would change my guidance or outlook on net interest income.
We would benefit significantly from rising interest rates.
Paul J. Miller - FBR Capital Markets & Co., Research Division
So the stable NII...
Doyle L. Arnold
I'd say the stable -- you're talking NII, you mean net interest income, yes?
Paul J. Miller - FBR Capital Markets & Co., Research Division
Yes.
Doyle L. Arnold
Yes, I tell you -- that's a comment if rate stay kind of as they are, low throughout the year.
Paul J. Miller - FBR Capital Markets & Co., Research Division
And then also with buying back your, I think, Series C preferreds, I believe towards the end of the year, so really you're not going to get any benefit from buying that back into 2014, am I correct?
Doyle L. Arnold
No. It's actually, it's a call.
It's not a buyback technically. It is callable at par September 15 of -- so the benefit will be there throughout the entire fourth quarter, whatever it is.
Paul J. Miller - FBR Capital Markets & Co., Research Division
And is that the only series that's callable? I was thinking of the slide presentation at Goldman Sachs, can you talk about 2 different series?
Doyle L. Arnold
Our trust-preferred Series B, which has $293 million of outstanding, is callable at par at any time. And that's 8% cost.
Paul J. Miller - FBR Capital Markets & Co., Research Division
So when you issue this new preferred, can you sort of call it back immediately or you just haven't set a time frame for that yet?
Doyle L. Arnold
I'm not going to give you enough detail, sorry, to specifically build a quarter-by-quarter model because I'm not going to get into the specific details of what was in my stress test and capital plan submitted to the Fed until I know they've completed their review and hopefully found that acceptable. But we'll announce what we need to announce, at this level of detail then.
Operator
Our next question will come from the line of Steven Alexopoulos with JPMorgan.
Steven A. Alexopoulos - JP Morgan Chase & Co, Research Division
Regarding the $200 million of series G announced today, I just wanted to follow up on that. It's a bit smaller than I was looking for as it relates to the Series C.
Doyle, what are you targeting for preferreds to represent as a percent of risk-weighted assets and just curious what your thoughts were to do this now versus closer to the timing when you could call the Series C?
Doyle L. Arnold
I'm sorry, I just blinked. What was the first part of your question?
James R. Abbott
Percentage of preferreds.
Doyle L. Arnold
Oh, yes. Yes, we've -- if you go back and look at our investor presentations throughout much of last year, I think starting with our hosted Investor Day in February we've, at various points in time, indicated that roughly 2% of risk-weighted assets would be our target level for preferred.
It's somewhere -- or other Tier 1, which would include today some trust-preferred. Including a trust-preferred that were today, I believe, at around 3.3%, something like that percent, so -- and the Series C makes up the bulk of it.
Trust-preferred is the second smallest piece and then series A and series E are the remainder of it. So we've also -- you didn't ask us specifically, but we've also indicated in various investor forum that we were unlikely to call the 800 -- approximately 800 million of Series E without issuing something on the order of 500 million to 600 million of perpetual preferred and partial replacement of it to get to those ratios.
And we've also noted that the trust-preferred long-term isn't of any use as Tier 1 capital. So that kind of gives you, I think, the context of where we're trying to get to and what the major levers are to pull to get there.
In terms of timing, I guess all I'd like to point out is that we've kind of consistently laid out a number of things that we want to do or need to address over the course of the next -- this year and the next to get the capital and financing structure in line with what they've indicated as our kind of medium-term target and to get the cost down. And we can't do it all at one time, so we're trying to basically trying to spread it out in a reasonable way.
Steven A. Alexopoulos - JP Morgan Chase & Co, Research Division
That's good color, Doyle. And just separately, I know the yield opportunity in the securities book isn't great, but it's better than the 28 bps that you're getting in the money market investments.
Are there any thoughts there moving some of that liquidity into the securities book as you try and stabilize net interest income?
Doyle L. Arnold
Yes. Something we look at continuously.
Again, we don't know how to spell NBS very well here. So I think we could -- the challenge is to get a 2% plus yield, you're going to take some kind of risk.
The risk of negative convexity is one that we're very wary of, doing very much of that. I think our actions back our words there.
But there are things between 25 bps and 200 basis points that we look at from time to time and they do a little of. We also have a much better read, I think as this quarter goes on, of just what the long-term-or at least medium-term deposit balances of our customers are.
As I mentioned, a significant chunk of that, $2.5 billion increase in deposits that occurred in the fourth quarter, has reversed already in the first quarter. So we'll have a little bit better idea of what we're looking at here as we get through the end of this quarter.
But we're not going to go take on a lot of negative convexity risk. Yes, we may deploy some of it into probably lower yielding securities, but that are very safe and don't have the risk of extending maturities or duration when you don't want them to.
I think that on previous call, David articulated very clearly what we're looking for. I laid down our criteria and invited any of you to call who had such an opportunity, and we're still waiting for the first phone call.
James R. Abbott
Like a March of Dimes telephoned around here. But maybe I'll just jump in those.
The amount that we might do in some of these securities that Doyle was referring to would be very small. I don't think it would really affect your net interest income outlook much.
Doyle L. Arnold
In any one pool, we might be talking about some hundreds of millions and then totally we might be talking about $1 billion to $2 billion if we can find that. But we want to keep a lot of our powder for loan growth, which we do believe will happen someday, maybe even starting this year.
Operator
Our next question comes from the line of Ken Zerbe with Morgan Stanley.
Ken A. Zerbe - Morgan Stanley, Research Division
Just a question in terms of the CPR assumptions on the CDOs. I think I heard that you said that they're now at 10%, which obviously is a pretty big jump from where they were.
Can you give us some context? How aggressive is that assumption?
Because I guess sort of, we thought 3% to 5% might be aggressive. Now it's 10%.
Obviously, this has been a big headwind for the OTTI or the losses on that portfolio historically, but I'm just trying to get a sense of the likelihood of further charges going forward?
Doyle L. Arnold
I'll offer a bit of commentary, and then invite David Hemmingway to chime in. We look at the actual and kind of historical trend every quarter.
And throughout most of -- throughout the latter part of 2011, probably first 2 or 3 quarters of 2012, then I mean, the number bounced around between under 2% and under 4%, and kind of was heavily anchored around 3%. So we don't think that number was aggressive.
We look -- we kind of look at the current -- what happened in the current quarter, what happened in the previous quarter and sort of what's the rolling fourth quarter average and [indiscernible] and just commentary on what we think is going on in the Street from investor bankers and other things. I don't think it was an aggressive assumption.
I think it was very real and anchored in observable data. What changed in the fourth quarter was the observable data where we had the actual prepayments in the fourth quarter were significantly higher than 10% --
Unknown Executive
12%.
Doyle L. Arnold
12%. But we didn't -- it's kind of hard to go from 3.5%, 2.5%, 4% to 3%, 12%, and say well maybe we ought to go to 15% or go to 12%.
So we moved it to 10% thinking that was a reasonable change based on what we observed and based on the fact that -- this change, by the way, only applied, remember, to the smaller banks, the ones that technically were not required under, I figure it was Basel or Dodd-Frank -- I think Dodd-Frank, to phase out trust-preferred, but for which the regulators proposed to phase them out over a 10-year period in lieu [indiscernible] that were issued early in the year or midyear but have not yet been made final. So David, you want to comment on anything else?
W. David Hemingway
I can say the 10% that you're referring to, while we raised it, even though the quarter was 12%, we raised it to 10%. It's only for 3 years, and then the model assumes that the prepayment rate drops back to 3%.
And really, over the next 3 years, you're going to get a lot of input and clarity as to what's going to happen in this portfolios because the 5-year deferral period is up. We believe there's going to be a lot of sales of banks and debit transactions.
I think probably 10% for 3 years is a reasonable assumption that we think once that these deferring banks get resolved, you could well see lower prepayments which is what's built into the model at the present time.
Doyle L. Arnold
And well within that time frame we should get a final rule from the Fed and others on whether or not smaller bank trust-preferreds will be phased out as Tier 1 Capital and, if so, over what time period.
Ken A. Zerbe - Morgan Stanley, Research Division
All right, that does help. And then just the other question I had, if you were to tender, do an early tender for your outstanding sub-debt, presumably you'd have to take all the, I guess, unamortized discount, I guess, as an accelerated amortization today, which would hit your equity by, I think, it's over $100 million or so.
Would -- is it -- should we assume, unless current replacement capital or an equity raise would be needed, to support that accelerated amortization from the early tender? Or is that presumably something you could ask for in your stress test by regulators as sort of a give me given your earnings over the next year?
Doyle L. Arnold
Well, again, I don't want to be dragged into what anything that's -- what's entered in our capital plan and stress test. I mean, you are correct in -- at the level of detail, at least, that you described that if you tendered and paid a premium to the current book value of debt, there would be an earnings and capital cost to doing so.
I think that I would just offer to comment that it would seem, unless we paid a premium that you just couldn't refuse, it would be unlikely that we would give anything close to a full tender of the sub-debt. And I would just say, I think that we would -- I mean, for me it would be a very bad trade to pay a premium for a sub-debt and issue -- have the trade be issuing common at below book value.
James R. Abbott
And unlikely that we'd put that kind of thing in our capital play.
Doyle L. Arnold
I realized that while you voted James #1, all of you, you demoted me this year. But I hope you don't think we're that stupid.
I didn't realize how far I'd plummeted [indiscernible].
Operator
Our next question comes from the line of Brian Foran with Autonomous.
Brian Foran
The 2% pref RWA target, I mean, I know you've said it over and over again. I think a lot of people probably were hoping you were being conservative.
And I guess maybe one of the things a lot of us struggle with, Wells, which is a much bigger bank has been very explicit about 1.5% RWA pref target. Some of the banks who are more in your size range are still kind of holding out to hope they can get away with like 75 or 100 basis points.
So why -- how should we think about -- do you think the whole industry eventually will be too? Or do you think your capital structure just will always be a little high on prefs.
How should we think about you relative to the rest of the industry?
Doyle L. Arnold
I would guess that if we look long term, 1.5% to 2% is not a bad place to be. The 2% is at the higher end of that range.
I do -- it's really hard to get a gauge. I mean, if you simply read the Basel rules, and we've got close to 10% CET1, there's no reason to have any preferred.
I mean, there's -- it absorbs no loss. I mean, I can understand your frustration.
But I think from the regulators, the FDIC standpoint, it doesn't absorb loss for our shareholders. It's not ongoing.
But it is a going concern, shouldering of lost to the FDIC. They're going to want to see it there.
I do think you're going to see a lot of preferred issue this year in replacement of trust-preferred or just as issuance. And I think -- the whole -- we're 3.3% today and the industry is at 1.1%.
We're going to come down, they're going to come up. And I think we'll all end up meeting somewhere between 1.5% and 2%.
It's not a -- my 2% number is not a, "Boy, I'm going to manage it every quarter to that number." But we're going to end up somewhere in that range.
James R. Abbott
One of the other advantages, Brian, is as you issue at these low rates you can fix in pricing. Presumably, loan growth and risk-weight assets growth will grow over time, so you can kind of grow into it.
That's another possible reason why you might want to set a little higher, a little sooner.
Doyle L. Arnold
Yes. That's why James is the #1 analyst.
Brian Foran
Fair enough. One follow-up on reserves and the potential for a negative provision in any given quarter.
I guess 2 related questions. One, kind of how do you think about where reserve to loans is likely to trend over time?
And then two, your reserve for unfunded commitment, I realize the provision is kind of a rounding error this quarter, but the stock is higher than most banks relative to your total outstanding unfunded commitments. So should we also think about that line item flipping to negative?
Or you mentioned that commitments are growing pretty fast right now and other stuff that's going on that would keep that at 0 even if the provision was negative?
Doyle L. Arnold
Well, I will just point out and you can decide whether or not to give us credit for 2 facts. One, we're -- among our peers, we had just about the lowest charge-off rate or one of the very lowest charge-off rates this quarter, and we've got one of the highest loan loss reserves.
Those 2 facts, and we've got one of the highest reserve on funding commitments relative. So I mean, you put all that together and it says there's very little pressure to boost reserves at this point.
Our models and our process are going to drive it. There's not a specific number we're targeting toward, but we're -- I just can't point to anything on the horizon that says that there's a reserve build in the future for the next 2 quarters, and probably continue to drawdown some reserve.
I can't find anything that points otherwise. One of my credit officers here staring at me from the other end of the table, but I'm a long way from him.
He can't reach me right now.
Harris H. Simmons
I might also just note that the recoveries in the fourth quarter were particularly strong. I mean, whether you have negative provisions or not, maybe driven largely by what recoveries continue to look like.
That's a predictable, vulnerable part of the equation. And that will be a big factor in our thinking.
James R. Abbott
We're going to probably end up going over time a little bit. So we'll maybe take the next 2 or 3 questions, but we will have to cut it off and call people back.
I apologize, there are a number of questions remaining.
Operator
Our next question will come from the line of Joe Morford with RBC Capital Markets.
Joe Morford - RBC Capital Markets, LLC, Research Division
I would just -- question looking closer at loan totals. I was wondering if there was any chance of getting a breakdown of kind of new credit extensions and advances versus the amount of paydowns and maturities and how those amounts compare with last quarter?
Doyle L. Arnold
I made James take that out of the call script because it was already too long, and he told me somebody was going to answer that question. He's the #1 analyst.
James R. Abbott
All right. We had -- production was up quite a bit.
It was up 16%, Joe, in the quarter and actually year-over-year. So if you're concerned about seasonality, we actually -- from a year-over-year perspective, it's just been a much stronger quarter.
Yields were -- on CNI, yields were up a little bit. Some of the other products were down slightly on production.
The yields are still dilutive to the existing book of business. But one little factor that I thought was interesting as I looked through all the data was that C&I -- the portfolio of C&I loans declined only 3 basis points in terms of yield this quarter compared to the third quarter.
That is a really slow rate of decline compared to what it's been dropping in the past. And so we are getting close to the point where the production is converging with the portfolio.
Joe Morford - RBC Capital Markets, LLC, Research Division
Okay, great. And then another quick question.
Just besides the comp reversal or the accrual there, are there any other outsized or unusual items in the fourth quarter expenses you'd quantify, preferred to some legal matters or OREO expenses tied to?
Harris H. Simmons
Well, OREO expenses doesn't evolve. I think we told you last quarter it's unlikely to be that low the following quarter.
Maybe -- but it's going to bounce around, it may go lower this quarter. Legal, go ahead.
Doyle L. Arnold
As you said, legal is probably up [ph]. I think the other noninterest expense line item was high and people have trouble reading that out.
Harris H. Simmons
Yes, and that's accrual for some -- I mean, about the cost of defending some cases and incurring for possible judgments or settlements. I'm obviously not going to get any more specific than that.
But it's hard to know what will happen in future quarters on some of these. We certainly think we've reserved adequately.
And if that turns out to be the case, that number will come back down because once you reserved it -- you don't reserve it, but if the future developments dictate, we may do it again. That's another one that's just hard to predict every quarter.
But I don't think, there's -- you should not take that increase from third quarter to fourth quarter and trend line it in the legal and in the other expenses.
Operator
We have time for one final questioner. Our final question will come from the line of Ken Usdin with Jefferies.
Kenneth M. Usdin - Jefferies & Company, Inc., Research Division
I just want to follow up on that prior question about loans. You had mentioned in your prepared remarks about how some of these quarters, some originations had paid down after the quarter end.
So just wondering, in that context, have you said that some of the deposits have gone out the door after the quarter? Can you give us some understanding of the -- on the loan side?
That temporary increase, was it meaningful or was it modest?
Doyle L. Arnold
I think I can just give you the fact. So where are we?
As of last Friday, so basically through the month of January, the $463 million increase in loans in the fourth quarter, most of which occurred in the last 4 to 6 weeks, was backed down by $100 million. Some categories still are growing like CMI.
So it's not a complete reversal. On the other hand, the deposits were up, I think $2.5 billion and about $1.5 billion of that, back down about $1.5 billion in a month.
Now we can't point to a long list of specific customers who drew down loans late in the year and then paid them back down. And we're not sure what that's all about, so those are the aggregate numbers we're talking about.
Kenneth M. Usdin - Jefferies & Company, Inc., Research Division
Okay. And then so when you net that altogether, if you presume that that carries forward, you have presumably average loans that are going to grow and then -- but the average size of the balance sheet should shrink so...
Doyle L. Arnold
Yes.
Kenneth M. Usdin - Jefferies & Company, Inc., Research Division
Shouldn't that help the margin? I mean, why -- are we still going to see margin pressure even underneath all of that?
And what would it be coming from that? Because I would think that your mix, your loan mix would be helping you get to that kind of NIM stability.
So where are you still seeing the compression coming from underneath?
Doyle L. Arnold
Well, I told you where it was coming from. But I do agree with you, the stated NIM will benefit significantly from -- if the size of the cash assets funded by deposits shrinks back down.
We've quantified in the prepared remarks that I think it was 5 basis points of margin compression was just -- you could just attribute it to the huge cash build-up during the quarter to the extent that reverses. That's a good offset to some of the other pressures.
So I didn't want to -- I don't want to overstate the NIM compression in any 1 quarter going forward. I come back to -- for that and a whole bunch of other reasons, I think the NIM compression will be much more modest this year than last.
If it occurs, most of it will be in the first half of the year, not that the last half of the year unless we see some dramatic change in loan pricing. And it will take half as much loan growth to completely offset it in terms of net interest income as we told you it would last year.
James R. Abbott
Yes, those [indiscernible] positives, I think.
Kenneth M. Usdin - Jefferies & Company, Inc., Research Division
Got you. And my last one is just on the characters -- I heard you walk through which lines you expect to grow within the loan buckets.
But where are you seeing this -- within the C&I book or within the books you expect to grow, what's changing underneath the surface? Like where is the end demand change the most to the positive within that commercial side of the portfolio?
Doyle L. Arnold
Energy.
Harris H. Simmons
Texas.
Doyle L. Arnold
Texas has done well. Have you heard of frac-ing.
I mean that's -- I mean -- by far the strongest is in Amegy, but it's pretty wide.
James R. Abbott
I was going to say, the energy loans are up about 5% over the last 6 months, not annualized. Finance, merger and acquisition related financing has also been strong.
It was not a strong -- came from behind in the last month, but it was a strong month, December for finance. So those are a couple of the categories.
Other categories are still doing well, but there are several -- there's only really one category that's down, which is construction-related businesses.
Doyle L. Arnold
We can come back. Let me just kind of come back to one thing.
One thing that gives me confidence that the decline, the run-up and then rundown in deposits was not TAG-related but also was kind of a onetime phenomena. We know of several very specific instances in which companies were sold late in the year and the deals closed in December because of an anticipated change in capital gains tax.
I mean, you've heard about some of the high-profile things about movie studios. Lucasfilms being sold and things of that nature.
But there are some closely held companies that we bank that -- and we got -- the companies were sold, the deal closed, a ton of cash came into their balance, to the owners of the company's balance sheet and then moved off fairly quickly. But there was a lot of "let's do this now before rates maybe rise" and of course, rates did rise subsequent to year end.
And I think that's what we're -- a lot of that $2.5 billion up and $1.5 billion down is -- can be tied to that. I believe we'll have to wrap it up, moderator.
Operator
Sure thing. I'll turn the call back over to management for any closing remarks.
James R. Abbott
Thanks, Siri, and thanks to all of you. We do have a list of those of you who are still waiting in the queue, and I will get back to you if you'll stay put.
We'll get back to you tonight. Thanks for standing by, and thanks for your attendance on the call today.
And we will see you at our conference sometime during the quarter.
Harris H. Simmons
Thank you much.
Operator
Thank you, gentlemen. Again, ladies and gentlemen, this does concludes today's conference.
Thank you for your participation and have a wonderful day. You may now all disconnect.