美國合眾銀行 (USB) 2022 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the U.S. Bancorp Third Quarter 2022 Earnings Conference Call. Following a review of the results by Andy Cecere, Chairman, President and Chief Executive Officer; and Terry Dolan, Vice Chair and Chief Financial Officer, there will be a formal question-and-answer session. (Operator Instructions) This call will be recorded and available for replay beginning today at approximately 11 a.m. Central Time.

I will now turn the conference over to George Andersen, Director of Investor Relations for U.S. Bancorp.

Thank you, Alan, and good morning, everyone. With me today are Andy Cecere, our Chairman, President and CEO; and Terry Dolan, our Vice Chair and Chief Financial Officer. During their prepared remarks, Andy and Terry will be referencing a slide presentation. A copy of the slide presentation as well as our earnings release and supplemental analyst schedules are available on our website at usbank.com.

I would like to remind you that any forward-looking statements made during today's call are subject to risk and uncertainty. Factors that could materially change our current forward-looking assumptions are described on Page 2 of today's presentation, in our press release and in our Form 10-K and subsequent reports on file with the SEC.

I will now turn the call over to Andy.

Thanks, George. Good morning, everyone, and thank you for joining our call. Following our prepared remarks, Terry and I will take any questions you have. And I'll begin on Slide 3.

In the third quarter, we reported earnings per share of $1.16, which included $0.02 per share of merger and integration charges related to the planned acquisition of MUFG Union Bank. Excluding these notable items, we reported earnings per share of $1.18 for the quarter. During the third quarter, we achieved record net revenue totaling $6.3 billion. Third quarter results were highlighted by strong revenue growth, well-controlled expenses and stable credit quality.

This quarter, we added $200 million to our loan loss reserve, reflecting loan growth and our consistent through-the-cycle underwriting approach to risk management. At September 30, our CET1 capital ratio was 9.7%. Our tangible book value per share totaled $20.73 at September 30 or 3.2% lower than the prior quarter driven by the impact of rising interest rates on our available-for-sale securities.

Slide 4 provides key performance metrics. Excluding notable items, we delivered a return on average assets of 1.24% and a return on average common equity of 16.2%. Our return on tangible common equity was 21.4% on a core basis.

Slide 5 highlights continued positive trends in digital engagement as consumer and business customers gain a deeper understanding of our digital capabilities and benefit from our digital-plus-human approach.

Slide 6 shows progress across our digital and payments initiatives that are both deepening our core competencies and expanding our competitive advantage, which we believe will drive meaningful profit and return differentiation to our company. One example of our digital initiatives is our launch of talech Register, a next-generation, all-in-one payments and business analytics platform that is helping to bring greater simplicity, convenience and efficiencies to the point of sale for our business banking customers.

On the right side of the slide, you can see the continuing momentum we are gaining in real-time payments transactions. Year-to-date through September 30, the total number of transactions are 17x higher than the full year 2020. Recently reintroduced an innovative, real-time payment solution to auto dealers, where we provide loan funds instantly after a loan contract is finalized. This gives our participating dealer clients greater control of their cash flow and helps to improve their day-to-day operational efficiency.

Turning to Slide 7. Our business banking initiative continues to gain traction with steady progress both in growing accounts and expanding wallet share. Growth in relationships with both banking and payments products has continued to meaningfully outpace growth in total relationships over the past 12 months. Looking ahead, we expect recently launched and innovative offerings in development to help further deepen our existing relationships between business banking and payments customers.

Now let me turn the call over to Terry, who'll provide more detail on the quarter.

Thanks, Andy. If you turn to Slide 8, I'll start with a balance sheet review followed by a discussion of third quarter earnings trends. Average loans increased 3.9% compared with the second quarter driven by 6.5% growth in commercial loans, 4.7% growth in mortgage loans and 6.0% growth in credit card balances. Commercial loan growth reflected increased business activity and higher utilization rates across both large corporate and middle market portfolios. Underlying demand remains healthy as we continue to focus on appropriate return opportunities to prudently deploy our capital.

In the retail portfolio, we saw solid linked quarter growth and year-over-year growth in credit card balances, reflecting strong spending activity and lower payment rates. Purchase mortgage market share gains and lower prepayment activity continued to support residential mortgage balance growth.

Turning to Slide 9. Total average deposits increased slightly compared to the second quarter. Growth in total interest-bearing deposits more than offset the impact of lower total noninterest-bearing deposit balances as customers respond to the rising interest rate environment. Total average deposits increased by 5.9% compared to a year ago.

Slide 10 shows credit quality trends, which continue to be strong across our loan portfolio. The ratio of nonperforming assets to loans and other real estate was 0.20% at September 30 compared with 0.23% at June 30 and 0.32% a year ago.

Our third quarter charge-off ratio of 0.19% improved slightly versus the second quarter of 2022 and third quarter of 2021 levels. The allowance for credit losses as of September 30 totaled $6.5 billion or 1.88% of period-end loans. The $200 million increase in our reserve this quarter was primarily reflective of loan growth and, to a lesser extent, uncertainties in the economic outlook.

Slide 11 provides an earnings summary. In the third quarter, we reported $1.18 per share, excluding $0.02 per share of merger and integration charges related to the planned acquisition of MUFG Union Bank.

Turning to Slide 12. Net interest income on a fully taxable equivalent basis totaled $3.9 billion, representing an 11.3% increase compared with the second quarter and a 20.6% increase from a year ago. Linked quarter growth was driven by strong earning asset growth and a 24 basis point increase in the net interest margin, which benefited from rising interest rates, partially offset by deposit pricing and short-term borrowing costs.

Slide 13 highlights trends in noninterest income. Noninterest income decreased 3.1% on a linked quarter basis as declines in mortgage banking and treasury management revenues were partially offset by stronger corporate payments revenue and an increase in other noninterest revenue. Compared with a year ago, noninterest income declined 8.3% primarily due to lower mortgage banking revenue, reduced deposit service charges reflecting changes in our policies and lower treasury management fees due to rising rates, partially offset by higher payments revenue and trust and investment management fees.

The decline in mortgage revenue primarily reflected lower refinancing activity in the market, which continued to pressure total application volumes and the related gain-on-sale margins given excess industry capacity. In the third quarter, total payments revenue increased by 4.9% compared with the year earlier.

Slide 14 provides linked quarter and year-over-year revenue growth trends for our 3 payments businesses. Because of the cyclical nature of our payments businesses, we believe year-over-year trends are a better indicator of underlying business performance in a normal environment. Credit and debit card fee revenue increased -- or decreased 0.5% on a year-over-year basis as the impact of higher credit card and debit card volume was more than offset by continued lower prepaid card activity.

Excluding prepaid card activity, which was elevated last year in connection with supporting unemployment programs, credit and debit card fee revenue would have increased 3.0% compared with the third quarter of 2021.

The bottom half of the slide illustrates the year-over-year growth rates in both merchant processing and corporate payments fee revenue over the last several quarters. Third quarter merchant processing revenue increased 3.6% year-over-year. Growth was negatively impacted by unfavorable foreign currency exchange rates given market volatility in Europe and specifically in the U.K. Excluding the FX impact, year-over-year growth in merchant fee revenue was approximately 9.4%.

Slide 15 provides some additional information on our Payment Services business. On the right side of the slide, you will see the continued strong momentum we are seeing in our tech-led revenue and partnerships within our merchant acquiring business. The key to that trajectory is the strong growth we have seen in new tech-led partnerships. Through the third quarter, new tech-led partnerships year-to-date were 2.5x the number of new partnerships we acquired in the entire year of 2019, and these partnerships are continuing to grow.

Turning to Slide 16. Noninterest expense increased 1.9% on a linked quarter basis, excluding merger and integration costs associated with the pending acquisition of Union Bank. The change in expense was driven by higher compensation, professional services and marketing and business development expenses.

Slide 17 highlights our capital position. Our common equity Tier 1 common -- our common equity Tier 1 capital ratio at September 30 was 9.7%.

On Slide 18, I'll provide some forward-looking guidance for U.S. Bank on a stand-alone basis. Again, this guidance does not include any potential impact of Union Bank. Let me start with full year 2022 guidance, which is consistent with our previous expectations. We continue to expect total net revenue to increase 5% to 6% in 2022 compared to 2021. We expect mid-teen growth in taxable equivalent net interest income, which has slightly improved from our previous outlook of low to mid-teens growth.

We continue to expect a decline in fee revenue for the full year, primarily due to the impact of higher interest rates on mortgage revenue due to lower refinancings in the market, lower deposit service charges due to pricing changes and a decline in other noninterest income.

We continue to expect positive operating leverage of at least 200 basis points in 2022, excluding the impact of merger and integration-related costs associated with the Union Bank acquisition. For the full year of 2022, we expect our taxable equivalent tax rate to be approximately 22%.

I'll now provide guidance for the fourth quarter. We expect both total revenue and total core expenses, excluding merger and integration costs, to increase by approximately 2% on a linked quarter basis. Net interest income will continue to be supported by earning asset growth and higher rates. However, our fee revenue will be lower, reflecting typical seasonality in some of our fee-based businesses.

Credit quality remains strong. Over the next few quarters, we expect the net charge-off ratio to remain lower than historical levels but to normalize over time. Changes in the allowance for credit losses near term will primarily reflect loan growth and changes in the economic outlook.

If you turn to Slide 19, I'll provide an update on our previously announced pending acquisition of Union Bank. In September of 2021, we announced that we had entered into a definitive agreement to acquire the core regional banking franchise of MUFG Union Bank. We continue to make significant progress in planning for the closing of the deal in the fourth quarter of 2022 while we await regulatory approval. As you know, regulatory approvals are not within the company's control and may impact the timing of the closing of the deal.

As a reminder, we expect to close on the deal approximately 45 days after being granted U.S. regulatory approval. As previously discussed, we are targeting a conversion date in the first half of 2023. The financial merits of the deal remain intact. Our EPS accretion estimates are unchanged, and we continue to estimate the acquisition will generate an internal rate of return of approximately 20%, which is well above our cost of capital.

The company's target CET1 capital ratio is 8.5%. Based on interest rates as of October 13, our CET1 capital ratio at close would approximately -- would approximate 8.3%. We expect the CET ratio to increase towards 9% as the purchase accounting valuation adjustments accrete into capital through earnings.

I'll hand it back to Andy for closing remarks.

Thanks, Terry. The advances we have made and continue to make across our business lines are paying off in terms of improved customer experience, new customer acquisition and deeper relationships. Expense management is a priority, and we continue to target positive operating leverage in 2022 and beyond.

We look forward to closing on the Union Bank acquisition pending regulatory approval. This is a unique opportunity at scale in one of our core markets, and we remain confident in the strategic and financial merits of the deal.

The credit -- the current credit environment is benign. In fact, our net charge-off ratio in the third quarter remain near historic lows, and we are not seeing any meaningful early-stage metrics that causes concern. That being said, we recognize the pressure points are building in several areas of the economy that could lead to stress in the future.

Borrowing costs are increasing, inflation is high, saving rates are starting to decline, and the stock market is well off its highs. So while the backdrop is favorable today, it would not be surprising to us to see an economic slowdown develop at some point driven by lower confidence levels, which may lead to reduced spending and business investment.

There are a number of scenarios that may play out over the next several quarters. We are preparing for a range of possible outcomes by prudently managing credit, liquidity and capital so that we continue to grow within our through-the-cycle risk management framework and deliver industry-leading returns.

In summary, our investments are paying off. We remain diligent stewards of capital. Our balance sheet is strong, and our focus is firmly fixed on managing the company for the long term.

I'll close by saying thank you to our employees. Your hard work and focus in doing the right thing for our customers and communities every day is recognized and appreciated.

We'll now open it up to Q&A.

(Operator Instructions) John Pancari with Evercore is online for a question.

On your commentary around the CET1 falling to 8.3% at close but need to get back up to the 9% before you resume buybacks, can you give us a little more color on that in terms of timing around when you think you can reach that 9% with the PA accretion and how we should think about the magnitude of buybacks at that time?

Yes. Thanks, John. So again, our expectation is a closing of 8.3%, but we do expect that, that will get to roughly 9% within about 4 quarters. And a big part of that will be, obviously, earnings growth driven by the accretion of the mark-to-market that will end up impacting earnings during that particular time frame. So our expectation and what we have been signaling in the past is that we would start our share buyback program once we get to that 9%. So about 4 quarters.

Okay. All right. That's helpful. And then just separately, can you just give us more color on the payments trends that you're seeing both in merchant processing and corporate payments businesses? And maybe if you could just talk about the potential moderation in activity there that you could see as the economy reacts to the Fed action.

Yes. John, this is Andy. And it's interesting, from a credit card spend standpoint, while the categories of spend, as we've talked about before, have shifted a bit, the level of spend is still fairly strong, about 10% on a year-over-year basis and about 30% above pre-COVID levels. We did get impacted, as Terry noted in his prepared comments, by the FX rate of the pound in our merchant processing. And excluding that, our revenues would have been up just over 9%.

We can -- we -- as well sort of a related topic, we see -- for 2 years, we have seen our consumer deposit levels by account increase. And as we mentioned last quarter, it's been flat. It's been flat for the second quarter and it's relatively flat, not down modestly, here in the third quarter. So we would expect some moderation of spend, but we're not seeing it yet.

Matt O'Connor with Deutsche Bank is online with a question.

This is Nathan Stein on behalf of Matt O'Connor. I just want to follow up quickly on the payments outlook. I think previously, you've talked about higher-than-normal year-over-year growth in the second half of this year before moderating to high single digits next year. Just to follow-up on what you were just talking about, does that guidance still make sense in this environment?

Yes. I mean our expectation is with respect to our payments revenues of high single digits. And we think that based upon what we are seeing with respect to consumer spend at this particular point in time that, that is still very realistic. As Andy said, relative to pre-COVID, volumes are continuing to be very strong. And while there is a bit of a shift with respect to where that spend is occurring, it's still continuing.

The other thing is that where -- if and when transaction levels start to tail off, I think the impact of inflation will have an offsetting effect. So we still feel pretty confident about it.

Okay. Great. That's helpful. And then one quick follow-up question. Can you just touch on the plan to manage the balance sheet post the deal closing, whenever that is? I think total assets were just above $600 billion at the end of this quarter. And with Union Bank, it will push you above the $700 billion threshold. So how are you thinking about that going forward?

Yes. Our expectation is that we'll continue to manage the balance sheet in a very prudent way, looking for and supporting relationships that are higher-profitable sort of relationships. Our expectation is that -- because I think where you're heading is really related to Cat II. Our expectation is that the earliest that we would get into that category is the end of 2024, but our ability to manage risk-weighted assets and balance sheet levels may extend that to some extent.

Mike Mayo with Wells Fargo is online with a question.

I think at a recent conference, you said that you will not achieve as much synergies from Union Bank next year as you originally thought, but no change in EPS guidance. Did I hear that correctly?

Yes, that is correct. And if you think about the dynamics, Mike, while the timing has changed and the synergies associated with the $900 million of synergies that we expect to achieve when fully implemented, the timing will be offset to some extent by the accretion of the mark-to-market will be a little bit stronger. So we still feel pretty confident with respect to the accretion levels that we have guided in the past.

Right. So instead of 75%, you said 40% to 45%?

Yes. 40% to 50%.

40% to 50% and 6% EPS-accretive. So if you're getting less merger synergies next year than you originally thought and you have the same EPS guidance, are you de facto increasing the end result of expected accretion from Union Bank?

Yes. I don't think that we're moving off of our estimates that we have provided in the past at this particular point in time. I think we need to get beyond the closing, the approval and then the closing before we, I think, would significantly change any estimates at this point.

Okay. And any early guidance for next year? Some other banks are giving some general kind of guidepost for 2023 in terms of operating leverage, and your guidance is saying this would be some of the best operating leverage in over 5 years, I guess. And just trying to figure out how confident you are in that continuing. I know you invested for a number of years in the tech infrastructure, and now you hope to capitalize on that at a time when business is coming back. But how much follow-through should investors expect for this positive operating leverage?

Yes. Mike, we're focused on positive operating leverage on a core basis, and we'll continue to achieve that. Next year has a fairly large moving part with Union Bank coming on. So that is going to be our top priority, is to make sure we successfully convert and integrate that in this company, which will allow for more positive operating leverage. But that -- we haven't provided any guidance update on that yet.

Erika Najarian from UBS is online with a question.

Terry, I'm going to have to ask you about the 8.3% pro forma CET1 that you mentioned you're estimating at close. I think I didn't realize that there has been a significant conversation in the market that perhaps has contributed to the 3-month underperformance with concern about the widening of the interest rate marks. So I think that at announcement, you were looking at a 50 basis point loan markup on UB's loan portfolio that you're bringing over.

Given that a significant portion of that portfolio is resi, I think there was a lot of math being done in the buy side that was, let's say, 100 basis points lower than what you were giving us. So I guess my question here is how is The Street getting the math wrong? Or are there protections that are built into the deal? Or are there hedges on UB's balance sheet that seems to be protecting your pro forma CET1 from a greater day 1 mark relative to how interest rates have moved since announcement?

Yes. I think it could be a combination of different things. I mean, for example, there is an expectation that they will deliver a certain level of tangible book value. And to the extent that their available-for-sale portfolio is impacted prior to the acquisition, there is kind of a make-whole provision within the agreement related to the available-for-sale securities. So that is part of it.

I think some of it could be just what the assumptions are related to the duration of the resi portfolio. And then again, we'll end up looking at and managing in this rate environment the balance sheet as prudently as we can to ensure that we are allocating capital to the appropriate businesses during this time frame. So it's a combination of a different -- a variety of different things, Erika.

Got it. So I guess that's the question -- as a follow-up question, that $6.25 billion that they have to deliver, there's a make-whole provision on the AFS portfolio but not on the loan portfolio, I'm presuming, is what you're saying, Terry?

That is true. Yes.

Got it. Is there -- and I'm presuming also there's a potential CDI offset at close that could help offset. But obviously, that would be a higher amortization cost in day 2 and after?

Yes. I mean it's a combination of a whole variety of things. I mean it's not only the interest rate marks, but it's what does the loan portfolio look like in terms of overall quality relative to kind of what our original estimates were. We think that, that is slightly better than CDI. I mean you have to look at all of the different moving parts in terms of coming up with what we think the final impact of the mark-to-market is going to be.

Bill Carcache with Wolfe Research is online with a question.

So we saw a nice remixing out of available-for-sale into held-to-maturity. Wanted to follow up on that and more specifically, whether there's room for you to take the mix of available-for-sale higher. It'd be helpful if you could just give some color on what guides your decision on how high the HTM mix can go.

Yes. Great question, Bill. So today, we're currently at about 53% held-to-maturity versus available-for-sale. And we moved some securities in early July when rates dipped that allowed us to be able to kind of increase that percentage.

The other thing that we take into consideration is the fact that with respect to the AFS portfolio, there's also floating rate securities that have very little impact with respect to AOCI. So when you end up combining that together, it's about 60% that is protected -- substantially protected from movements in interest rates. So think about it almost as a 60-40 sort of split in terms of the sensitivity to interest rates.

And is there opportunity? I mean we'll continue to look at whether that makes sense relative to our balance sheet positioning. But at least right now, we feel pretty comfortable with that mix. It gives us the AOCI protection as well as some flexibility with respect to managing the balance sheet.

And then the other thing you have to kind of think about is that we have Union Bank that hopefully we'll be closing on here in the fourth quarter. And that ends up influencing decisions that we make with respect to the positioning of the balance sheet and -- excuse me, of the securities portfolio. And then also with respect to deposits, we're going to have a significant amount of deposits coming on, and that's going to help us as well.

Very helpful. And then separately, I wanted to follow up on funding and more specifically, the decrease in noninterest-bearing deposits. Is it reasonable to expect that your mix of noninterest-bearing deposits is going to gradually revert to pre-COVID levels as the Fed proceeds with QT? Maybe if you could address that, if you have any view sort of at the industry level and then specifically for USB.

Yes. Well, I certainly think that QT is going to put pressure on deposits overall. And as interest rates rise, you'll continue to see a shift out of noninterest-bearing toward interest-bearing. Whether -- I mean, I think that the assumption with respect to pre-COVID may make sense depending upon how much they bring their balance sheet down, et cetera, how much liquidity they leave in the system.

With respect to us, it probably -- if I had to make estimates, I would say that it probably stays a little bit better than where it was pre-pandemic in part because we've grown our consumer portfolio, which tends to offset that. And then we've also worked hard to focus within our corporate trust business on retaining those deposits that are more operational in nature. And as -- because they're more operational in nature, they tend to be more noninterest-bearing as opposed to interest-bearing.

That makes sense. And sticking with that theme, in terms of the mix of debt relative to your overall funding, that's still well below fourth quarter '19 levels. Can you speak to the likelihood that we should expect that to gradually remix to -- back to pre-COVID levels?

Yes. Again, I think the broad assumption would be fair. But again, keep in mind that Union Bank is going to be coming on. And they have a significant -- they actually have a higher percentage within their deposit base that's consumer-based. And so I think you have to kind of look at the entire mix. And I think we're going to be -- U.S. Bank, I think, specifically is probably going to be in a better position than we were pre-pandemic.

That's great. Finally, if I could squeeze in one last one. Could you speak to your ability to bring off-balance-sheet money market funds back on balance sheet, maybe some of the dynamics surrounding that?

Yes. Great question. So currently, we have about -- roughly $130 billion in money market funds. And that is a business that's very tied to our institutional investor services business. Based upon what we need to do from a funding perspective, we have the ability from a pricing point of view to bring that back on balance sheet or off balance sheet, but it's pricing decisions that then influence customer behavior. And it is a great source for us from a funding standpoint as there is more pressure on deposits going forward.

Gerard Cassidy with RBC is online with a question.

Terry, I think in your comments, you talked about the residential mortgage business, and the gain-on-sale margins were lower. And I think you referenced there's still excess capacity in that line of business. Can you share with us your outlook? Are you seeing some of that capacity coming out in the fourth quarter if rates remain elevated for the refi activity, of course, being negatively affected by elevated long-term rates?

Yes. We certainly are seeing capacity come down in the industry. I think you've seen various announcements of bank and nonbank mortgage operations kind of pulling back or reducing capacity. I think it probably still needs to reset to some extent, what's happening with rising rates and the impact that it might have on both refinancing as well as home sales, but we do see it coming down.

Our expectation is that gain-on-sale margins stabilize and, probably for us, improve a bit. In the third quarter, our gain-on-sale was down maybe a little bit more than what we had anticipated. But about half of that is really driven by the mix between correspondent and retail. And the other half of it is really more revenue recognition, some timing issues and some secondary kind of market valuation issues, which we think will reverse a bit.

Very good. I saw when we looked at your average loan portfolio, obviously, you had some very nice growth in C&I, very strong growth year-over-year. We're hearing some market chatter, and I don't know if it's just the equity REITs are squawking. But apparently, the commercial real estate mortgage business might be a little more or less liquid today. Any color that you guys can offer on what you're seeing in commercial mortgage real estate? Is it an area that you're pulling back some? I know you had growth, but maybe some color and insights on what you're hearing and seeing and what your views are for that business?

Yes. Certainly, with respect to commercial real estate, it's an important business for us and one that we'll continue to be very focused on. I think part of it is just dynamics in the marketplace. If you end up looking at the large corporate REIT sort of financing, I -- we continue to lend in that particular space. We actually probably saw some growth in the third quarter.

But we do know homebuilders are starting to pull back and even making decisions to maybe pull out of potential projects that they were planning on doing. And then in the middle market space in terms of commercial real estate, it's actually still from a pricing standpoint pretty competitive and people extending terms associated with that. And that's the space that we're not comfortable with at this particular point in the cycle. So we're seeing a little bit of pressure maybe in the middle market space, but we're fine with that.

Ken Usdin with Jefferies is online with a question.

Terry, I wanted to just ask a little bit more on the whole outlook for deposits. I know you talked about the mix. Just wondering what you guys are seeing given the fact that you have a little bit of a different mix versus most of the peers in terms of -- any updates on what you're thinking about where betas eventually go to, just given that we're in a completely different rate regime than we might have all thought 6 months ago?

Yes. Well, maybe I'll kind of start just by reiterating if we end up looking at the mix of our deposit base, it has changed a little bit relative to maybe what we were 3, 4 years ago, et cetera. Certainly, when we went through the last rate rising cycle, a little stronger consumer, a little more focus around operational deposits within our trust business, et cetera.

Trust is also -- on a relative basis represents only about 15% of our deposit base today. And then again, think about Union Bank coming on and some of the dynamics that, that will bring. But coming back to deposit betas then maybe with that context, we're actually performing probably better than what we might have expected. I think deposit betas were about -- in the second quarter, about 30% coming up from about 20%. We would expect that to be maybe in the high 30s as we get into the fourth quarter.

But certainly, as rates continue to rise and if Fed continues to be very aggressive, it's going to migrate to higher percentages. Our through-the-cycle estimate right now is still kind of mid-30s, but we kind of think about the entire rate cycle that we're going through.

Okay. And then as a follow-up to that, so as that plays forward, you're still expecting good NII growth into the fourth quarter. Can NII -- I know you're not giving '23 guidance, but just can you help us understand like what is the gives and takes to make -- can you still grow NII post the fourth quarter sequentially? Or is there just a reality check about some of these mix and beta functions that make that tougher?

Yes. Well, I think that we can continue to grow NII as we get beyond the fourth quarter. What I would say, though, is that probably the pace of growth changes in the industry, and that will be a function of -- it will be a function of the deposit betas changing. And the loan growth that certainly the industry has been experiencing, I think, is very strong. I don't know whether it will keep that particular base as we kind of continue to move forward.

Betsy Graseck with Morgan Stanley is online with a question.

On your own securities [portfolio], it would be helpful -- maybe you could give us a sense as to how we should be thinking about the pull to par in the AFS book. Like if -- I know it's a big if, but if rates don't change from here, how many quarters or years is it that we should be baking into our models?

Yes. Well, I think what we have said is that the duration of the portfolio is a little over 5 years. So if you think about that, I think that probably gives you some guide with respect to how AOCI kind of comes into the capital equation.

Okay. So it should -- it's like a longer than 5-year period?

Yes. A little bit longer but not much.

Okay. And is there anything you're thinking about with regard to restructuring the books once Union Bank comes on?

Yes. It's -- again, we'll take a look at all of the balance sheet and kind of see where it makes sense. We do expect that we might remix the securities portfolio when it comes on. We may end up increasing HTM as a result of what we bring over, et cetera. So it is definitely something we will take a look at.

And then just separately on the consumer lens that you have in the payment side. Could you just give us a sense as to what you saw in 3Q? I know 3Q tends to be a relatively strong quarter for you. Any changes in behavior? Any insights as to how you think 4Q will end up shaping up?

So Betsy, the consumer spend levels overall continue to be strong. They were up -- and credit card spend about 10% on a year-over-year basis, about 30% above pre-COVID levels. The mix of categories has changed a bit, consistent with what we've talked about before, a little bit away from hard goods to more services and a little bit away from nondiscretionary -- to nondiscretionary from discretionary. So that shift in spend that we have been discussing continues, but the overall levels are also still very strong.

And the savings levels of the consumers are hanging in there?

Yes, Betsy. So I mentioned that we saw increases across all stratas of balances for about 2 years, but the last 2 quarters have been stable across all the categories. So not growing anymore, but not shrinking dramatically as well.

Ken, I just want to clarify something with respect to deposit betas just to make sure that I've got the percentages right. The deposit betas in the second quarter were about 20%. Deposit betas in the third quarter were just shy of 30%. And we would expect that increase to continue as rates rise and to accelerate a bit, and again, probably in the high 30s when we get into the fourth quarter.

We have no further questions at this time. I will now turn it back to George Andersen. Please continue.

Thank you for listening to our earnings call. Please contact the Investor Relations department if you have any follow-up questions.

This concludes today's conference. Thank you for participating. You may now disconnect.