Capital One Financial Corp
NYSE:COF
Earnings Call Analysis
Q4-2023 Analysis
Capital One Financial Corp
The company continues to heavily invest in marketing, focusing on enhancing its franchise, including increased media spend and promotions for its travel portal, airport lounges, and Capital One Shopping. Despite pulling back in auto origination to 7% below the previous year, the firm shows resilient growth, especially in federally-insured consumer deposits, indicating a strong digital-first strategy. Although consumer banking revenue fell by 17% due to lower auto loan balances and higher deposit costs, the company's long-term strategy highlights strong growth opportunities across its business, driven by technology transformation and data utilization for better customer solutions. The company is keen on expanding its digital banking presence nationally, utilizing its technology and branding investments.
Auto loan charge-off and delinquency rates showed increases both year-over-year and sequentially, which the company attributes to typical seasonal patterns. Nevertheless, auto credit has begun to stabilize. Despite a 1% decrease in loan balances, opportunities for improving margins have become more favorable due to recent interest rate reductions. Moreover, they are cautiously optimistic about the auto lending market, emphasizing disciplined pricing strategies and comfort with the terms set. In the commercial banking sector, there was a strategic choice to reduce less attractive deposit balances, thereby improving the average rate paid on these deposits. Charge-off rates increased modestly, yet commercial banking credit trends are under pressure, particularly within the commercial office portfolio.
The company has made a concerted effort to improve its annual operating efficiency ratio by 99 basis points in 2023, outperforming their modest improvement expectation. This has been achieved through a focus on revenue growth and stringent cost management, propelled by technological investments. The balance sheet has been fortified, with liquidity and capital being added.
The proposed rule from the Consumer Financial Protection Bureau (CFPB) on late fees, which could reduce these fees by approximately 75%, is expected to bring significant short-term impacts on profits. The company is planning to mitigate these impacts with policy changes, product adjustments, and strategic investments. Although some mitigations have already started, the bulk of the response actions will follow the implementation of the new rule, projected to take effect in the second half of the year following anticipated industry litigation.
You don't have any saved screeners yet
Good day, and thank you for standing by. Welcome to Capital One Q4 2023 Earnings Call. [Operator Instructions] Please be advised that today's conference is being recorded.
I would now like to hand the conference over to your speaker today, Jeff Norris, Senior Vice President of Finance. Please go ahead.
Thanks very much, Amy, and welcome, everyone, to Capital One's Fourth Quarter 2023 Earnings Conference Call. As usual, we are webcasting live over the Internet. To access the call on the Internet, please log on to Capital One's website at capitalone.com, and follow the links from there.
In addition to the press release and financials, we have included a presentation summarizing our fourth quarter 2023 results.
With me today are Mr. Richard Fairbank, Capital One's Chairman and Chief Executive Officer; and Mr. Andrew Young, Capital One's Chief Financial Officer. Rich and Andrew are going to walk you through the presentation.
To access a copy of the presentation and the press release, please go to Capital One's website, click on Investors, then click on Quarterly Earnings Release.
Please note that this presentation may contain forward-looking statements. Information regarding Capital One's financial performance and any forward-looking statements contained in today's discussion and the materials speak only as of the particular date or dates indicated in the materials. Capital One does not undertake any obligation to update or revise any of this information, whether as a result of new information, future events or otherwise. Numerous factors could cause our actual results to differ materially from those described in forward-looking statements. And for more information on these factors, please see the section titled Forward-looking Information in the earnings release presentation and the Risk Factors section of our annual and quarterly reports accessible at Capital One's website and filed with the SEC. Now I'll turn the call over to Mr. Young. Andrew?
Thanks, Jeff, and good afternoon, everybody. I'll start on Slide 3 of today's presentation.
In the fourth quarter, Capital One earned $706 million or $1.67 per diluted common share. For the full year, Capital One earned $4.9 billion or $11.95 per share. Included in the results for the fourth quarter was a $289 million accrual for our current estimate of the FDIC special assessment. Net of this adjusting item, fourth quarter earnings per share were $2.24 and full year earnings per share were $12.52.
On a linked-quarter basis, Growth in our Domestic Card business drove period-end loans up 2%, and average loans up 1%. Period-end deposits increased 1% in the quarter and average deposits were flat. Our percentage of FDIC insured deposits grew to 82% of total deposits in the fourth quarter.
Revenue in the linked quarter increased 1% driven by both higher net interest and noninterest income. Noninterest expense was up 18% in the quarter. Operating expense increased 15%, with roughly half of that increase driven by the FDIC special assessment. The full year operating efficiency ratio, net of adjustments, improved 99 basis points to 43.54%. Provision expense was $2.9 billion, comprised of $2.5 billion of net charge-offs and an allowance build of $326 million.
Turning to Slide 4. I will cover the allowance balance in greater detail. The $326 million increase in allowance brings our total company allowance balance up to approximately $15.3 billion as of December 31. The total company coverage ratio is now 4.77%, up 2 basis points from the prior quarter, largely driven by a higher mix of card assets.
I'll cover the drivers of the changes in allowance and coverage ratio by segment on Slide 5. Outside of interest rates, most of our economic assumptions are largely unchanged from the third quarter, and we continue to assume several key economic variables modestly worsened from today's levels.
In our Domestic Card business, the coverage ratio decreased by 16 basis points to 7.63%. The allowance balance increased by $336 million. The predominant driver of the increased allowance was the loan growth in the quarter.
In our Consumer Banking segment, the allowance was essentially flat at roughly $2 billion. Coverage increased by 4 basis points to 2.71%, driven by a decline in auto loans in the quarter. And finally, in our Commercial Banking business, the coverage ratio declined by 3 basis points to 1.71%. The allowance decreased by $37 million, primarily driven by the charge-offs of office real estate loans in the quarter.
We have included additional details on the office portfolio on Slide 17 of tonight's presentation.
Turning to Page 6. I'll now discuss liquidity. Total liquidity reserves in the quarter increased by $2.3 billion to about $121 billion. The increase was driven by a higher market value of our investment securities portfolio, partially offset by modestly lower cash balances. Our cash position ended the quarter at approximately $43.3 billion, down $1.6 billion from the prior quarter. You can see our preliminary average liquidity coverage ratio during the fourth quarter was 167%, up from 155% in the third quarter. The increase in the LCR was driven by holding more of our cash balances at the parent company versus our banking subsidiary.
Turning to Page 7, I'll cover our net interest margin. Our fourth quarter net interest margin was 6.73%, 4 basis points higher than last quarter and 11 basis points lower than the year ago quarter. The quarter-over-quarter increase in NIM was largely driven by a continued mix shift towards card loans and higher asset yields, partially offset by higher rate paid on deposits.
Turning to Slide 8. I will end by discussing our capital position. Our common equity Tier 1 capital ratio ended the quarter at 12.9%, approximately 10 basis points lower than the prior quarter. Asset growth, common and preferred dividends and the share repurchases more than offset net income in the quarter.
And with that, I will turn the call over to Rich. Rich?
Thanks, Andrew. Good evening, everyone. Slide 10 shows fourth quarter results in our credit card business. Credit Card segment results are largely a function of our Domestic Card results and trends, which are shown on Slide 11.
Top line growth trends in the Domestic Card business remained strong, even with growth moderating somewhat in the fourth quarter. Purchase volume for the fourth quarter was up 4% from the fourth quarter of last year. Ending loan balances increased $16 billion or about 12% year-over-year. Average loans increased 14%. And fourth quarter revenue was also up 14% year-over-year, driven by the growth in purchase volume and loans.
The charge-off rate for the quarter was up 213 basis points year-over-year to 5.35%. The 30-plus delinquency rate at quarter end increased 118 basis points from the prior year to 4.61%.
On a sequential quarter basis, the charge-off rate was up 95 basis points, and the 30-plus delinquency rate was up 30 basis points. For the month of December, the charge-off rate was 5.78%, including a onetime impact of 15 basis points described in a footnote in the monthly credit 8-K. Adjusted for this impact, the monthly charge-off rate for December would have been 5.63%.
Pulling up on Domestic Card credit, we believe that normalization has run its course and credit results have stabilized. The 30-plus delinquency rate has been stable on a seasonally adjusted basis for a number of months now. Since August, our monthly delinquency rate has been moving in line with normal seasonality, and that stable ratios relative to the same month in 2018 and 2019. And at this point, we have a pretty good window into January as delinquency entries in December indicate continuing delinquency rate stability in January.
We've always said that delinquencies are the leading indicator of where charge-offs are going. Charge-off rate tends to follow delinquency rate by about 3 to 6 months. Based on the stability we've seen in our delinquencies since August and extrapolating from our current delinquency inventories and flow rates, we believe the charge-off rate is stabilizing now and settling out to about 15% above 2019 levels. I give this window because investors have been asking for quite some time, when will charge-offs level off? So this is the point where we see that happening, meaning charge-offs should move more or less with seasonality in the coming months. This window comes from modeling the flows in our delinquency buckets, which have stabilized, and our recoveries, which have also stabilized and started to rebuild.
This isn't designed to be longer-run guidance, but rather to indicate that charge-offs are finally moving more or less with seasonality over the near term. In the longer run, there could be additional forces such as potential pressure from economic worsening, and potential benefits, from the depletion of deferred charge-offs from the pandemic and recoveries picking up over time from increased inventories.
Noninterest expense was up 11% compared to the fourth quarter of 2022, with increases in both operating expense and marketing expense. Total company marketing expense of about $1.25 billion for the quarter was up 12% year-over-year.
Our choices in our card business are the biggest driver of total company marketing. We continue to see attractive growth opportunities in our Domestic Card business. Our opportunities are enhanced by our technology transformation. Our marketing continues to deliver strong new account growth across the Domestic Card business. And in the fourth quarter, marketing also included higher media spend and increased marketing for franchise enhancements like our travel portal, airport lounges and Capital One Shopping. We continue to lean into marketing to drive resilient growth and enhance our Domestic Card franchise. As always, we're keeping a close eye on competitor actions and potential marketplace risks.
Slide 12 shows fourth quarter results for our Consumer Banking business. In the fourth quarter, auto originations declined 7% year-over-year, driven by the decline in auto originations, Consumer Banking ending loans decreased about $4.5 billion or 6% year-over-year. On a linked-quarter basis, ending loans were down 2%.
We posted another strong quarter of year-over-year growth in federally-insured consumer deposits. Fourth quarter ending deposits in the consumer bank were up about $26 billion or 9% year-over-year. Compared to the sequential quarter, ending deposits were up about 2%. Average deposits were up 11% year-over-year and up 1% from the sequential quarter.
Powered by our modern technology and leading digital capabilities, our digital-first national direct banking strategy continues to deliver strong consumer deposit growth and gradually increase the percentage of total company deposits that are FDIC insured.
Consumer Banking revenue for the quarter was down about 17% year-over-year, largely driven by lower auto loan balances and higher deposit costs. Noninterest expense was down about 3% compared to the fourth quarter of 2022. Lower operating expenses were partially offset by an increase in marketing to support our national digital bank.
The auto charge-off rate for the quarter was 2.9% and up 53 basis points year-over-year. The 30-plus delinquency rate was 6.34%, up 72 basis points year-over-year. Compared to the linked quarter, the charge-off rate was up 42 basis points, while the 30-plus delinquency rate was up 70 basis points. Both of these linked quarter increases were in line with typical seasonal expectations.
Monthly auto credit began to stabilize even earlier than Domestic Card credit results. On a monthly basis, auto delinquency rate and charge-off rate have been tracking normal seasonal patterns since the first half of 2023 and continued to do so through December.
Slide 13 shows fourth quarter results for our Commercial Banking business. Compared to the linked quarter, ending loan balances decreased about 1%. Average loans were also down about 1%. The modest declines are largely the result of choices we made earlier in the year to tighten credit.
Ending deposits were down about 9% from the linked quarter. Average deposits were down about 7%. The declines are largely driven by our continuing choices to manage down selected less attractive commercial deposit balances. Reducing these less attractive deposits also drove the 14 basis point linked quarter improvement in our average rate paid on commercial deposits. Fourth quarter revenue was down 5% from the linked quarter. Noninterest expense was also down about 5%. The Commercial Banking annualized charge-off rate for the fourth quarter increased 28 basis points from the third quarter to 0.53%. The Commercial Banking criticized performing loan rate was 8.81%, up 73 basis points compared to the linked quarter. The criticized nonperforming loan rate was down 6 basis points to 0.84%. Commercial Banking credit trends were largely driven by continuing pressure in our commercial office portfolio.
Slide 17 of the fourth quarter 2023 results presentation shows additional information about the remaining commercial office portfolio, which is less than 1% of our total loans. In closing, we continued to deliver solid results in the fourth quarter. We posted another strong quarter of top line growth in Domestic Card revenue, purchase volume and loans. Domestic Card and auto delinquency trends were in line with normal seasonal patterns, a continuing indicator of stabilizing consumer credit results. We grew consumer deposits and total deposits. And we added liquidity and maintained capital to further strengthen our already strong and resilient balance sheet. Our annual operating efficiency ratio net of adjustments for the full year 2023 was 43.54%. In 2023, we saw incremental opportunities and made choices to grow revenue and tightly manage costs to achieve a 99 basis point improvement in our annual operating efficiency ratio. The actual improvement was better than the "modest improvement" we had been expecting. Over the last decade, we've driven significant operating efficiency improvement even as we've invested to transform our technology. And we continue to drive for efficiency improvement over time. For the full year 2024, we expect annual operating efficiency ratio, net of adjustments, will be flat to modestly down compared to 2023. Our expectation includes the partial year impact of the proposed CFPB late fee rule, assuming that rule takes effect in October 2024. Pulling way up, our modern technology capabilities are generating an expanding set of opportunities across our businesses. We continue to drive improvements in underwriting, modeling and marketing as we increasingly leverage machine learning at scale. And our tech engine drives growth, efficiency improvement and enduring value creation over the long term. We remain well positioned to deliver compelling long-term shareholder value and to thrive in a broad range of possible economic scenarios.
And now we'll be happy to answer your questions. Jeff?
Thank you, Rich. We'll now start the Q&A session. [Operator Instructions] Amy, please start the Q&A.
[Operator Instructions] And our first question comes from the line of Sanjay Sakhrani with KBW.
And Rich, thank you for the color on the charge-offs. I know you cited the pluses and minuses from here -- from stabilizing charge-offs. But I'm curious if you feel like the consumer positioning leans towards improvement here as inflation declines, real income growth has resumed and should interest rates come down. This, coupled with the recoveries could benefit the charge-off rate, correct? And I'm just thinking sort of how to think about the reserve rate going forward.
Yes. So Sanjay, yes, so first of all, my comments, I just -- I wanted -- with my credit comments and my sort of extrapolated look at our delinquency buckets, I really wanted to share where our charge-offs are settling out, which is about 15% above 2019 levels. And we should also note, by the way, that in any year, first half losses are seasonally higher than second half losses. So within any normal year, the first half, is the peak. Now as I said, we're not really giving longer run guidance, but let's think about the dynamics about how things could go from here. From an economy point of view, we're certainly in a strong economy, unemployment, at a pretty remarkable place. So I think, if anything, unemployment could have more downside than it has upside. Perhaps inflation has more upside than it does downside, but I don't have any more insight to those than anybody else does. I think I also would want to kind of reinforce your point. There are two good guys that should play out over time. And we've been talking for a long time about the delayed charge-off effect from the pandemic. And when you think about that, the pandemic had set you just absolutely unusual experience for consumers with all of the stimulus and the forbearance and so on, that we certainly have believed that charge-offs that were otherwise going to happen at that time some got averted permanently, but I think some got delayed. And so this phenomenon that we call delayed charge-offs, I think it's not a really quantifiable effect, but I think it's very much been a part of what's happening with -- in the normalization and something that intuitively will run its course. And the other thing is recoveries. So recoveries, our recovery, while the rate per charge-off dollar remains strong, the number of charge-off dollars in inventory, thanks to the pandemic, were at a really depressed level. And so we have bottomed out there and they're starting to -- now inventories are starting to increase, of course, as credit has normalized, and that should also gradually be a good guy. Also when I look at our origination strategy and the underwriting choices we make, these are consistent with longer-term losses that are lower than where we are now. So we consciously sort of focused our credit commentary to really focus on where things settle out. And then there's a list of forces that could work in either direction, but I think you certainly point out some of the good guys.
Okay. Just a follow-up. Maybe, Andrew, could you just talk about the reserve rate on a go-forward basis and how we should think about it? Should it stabilize? Can it come down? And maybe just also on the NIM, with rates coming down, how we should think about the movement over the course of the year?
Sure, Sanjay. Let me compartmentalize those two things. The NIM will be a whole separate answer. But with respect to allowance, well, let me first start with just a tactical housekeeping item, which is a reminder that in Q4, we have some seasonal balances that quickly pay off in the first quarter and therefore, have negligible coverage. So the coverage ratio in Q4 is modestly lower as a result of that dynamic and it reverses itself in Q1, but again, a real modest effect there. Longer term, though, projected losses are really going to be the biggest driver of coverage. And as we've said before, delinquencies are the best leading indicator of that, and Rich just provided a fulsome description of all of the forces at play there. So from a reserve perspective, every quarter, we're just going to be looking at the next 12 months of projected losses with the first 6 more consequential in the calculation but also far more predictable given the visibility that we have through delinquencies. And then the remainder of that window really informed by economic assumptions. And then the reversion to the long-term average. And so over the last few quarters, things have played out consistent with or slightly better than what we've expected, and you've seen the coverage ratio in card roughly stay flat.
So I think it's important to note that even in a period where projected losses in future quarters are lower than today and might otherwise indicate a release. We could very well see a coverage ratio that remains flat or only modestly declined as we incorporate some of that uncertainty into the allowance, but eventually, in a scenario like that after a period of coverage stability like we've seen, you would see coverage coming down in the release of nongrowth-related reserves.
NIM?
NIM? Sure. There's a lot of factors at play with NIM, and maybe I'll do the same housekeeping with NIM, just to remind everyone that in the first quarter, with one fewer day, we're going to see roughly a 7 basis point headwind there. But let me then also enumerate here the puts and takes to NIM.
On the tailwind side, growth in card balances as a percentage of the balance sheet and even within those balances, possibly higher revolve rate, certainly a tailwind, and that's something we've seen over the last couple of quarters, and then also a lower cash balance. We've talked about this before, but cash balances today in total of $43 billion. I think the number is, with about $37 billion at the Fed, is quite a bit higher than prepandemic. I don't think we'll get back to where we were prepandemic, but I would expect over time that will come down from today's level. So that would also be a tailwind to NIM. On the headwind side, even though the Fed has stopped moving up in July, we continue to see some deposit product rotation and it creates a bit of upward pressure to the deposit betas. And then even if the Fed starts decreasing rates, we're going to see the assets reprice more quickly than the deposits, and the competitive environment and the backdrop of QT will potentially have an impact on betas on a downward cycle. So that would create a bit of margin pressure. And then a couple of other things that I would just highlight as potential headwinds, the uncertainty around potential regulatory changes that could impact interest income as well as just the path of credit, you've seen suppression go up over the last few quarters as losses has go up. So that also creates some pressure to NIM.
So I know that, that was a list of puts and takes, but I partially go into that level of detail to say, it's kind of hard to say where NIM is going to go in the near term, especially because the path of interest rates remains fairly wide at this point, but kind of gives you a sense of all of the forces at play. But over the much longer term, I would say there's nothing really structurally different about our balance sheet from where it was pre-pandemic that leads me to believe that NIM will be materially different than where it was based on at least what we know today.
Our next question comes from the line of Ryan Nash with Goldman Sachs.
Rich, when I look, you put about $4 billion of marketing expense for the second straight year, and you're continuing to drive strong growth. We're hearing about some others with a little bit of a more cautious tone on growth. So maybe just talk about, are you leaning in? And do you expect marketing to increase? And really, where are you seeing the best opportunities in the market? And how are you thinking about growth looking ahead? And I have a follow-up.
Okay. Thanks, Ryan. We feel very good about the opportunities in the marketplace. So we are leaning in. We're definitely leaning in the -- you can see, obviously, there was quite a lot of marketing in the fourth quarter. But we continue to see opportunities across the board, especially in the card business. But so just pulling up, there are a few key factors driving our marketing that we want to continue to emphasize. First of all, we're just really excited about the growth opportunities across our business. We're making -- we have, over the last number of years, made some what we call sort of adjustments around the edges and trimming around the edges.
Lately, there's really even not a lot of trimming around it, it's just we're in a very sort of stable place with respect to the business we're going after, the results we're getting and the zeal that we have to capitalize on that. And I think our technology transformation, of course, has really been beneficial, but it allows us to leverage more data and more machine learning models to identify basically more attractive opportunities for investments and to create better and more customized solutions for customers along the way. So just the overall opportunities continue to be very strong.
The second part of our marketing investment, of course, relates to our quest to win at the top of the market. And we've been going after heavy spenders now for almost 15 years. And that needs sustained, high levels of investment, and you can see those out in the marketplace in flagship products, in groundbreaking experiences like -- things like best-in-class digital customer experiences, really high-level, elite customer servicing, online travel portal. And in the intersection of risk management and the quest to go to the top of the market, the incredible importance of advanced fraud defense is to ensure that the card always works.
And increasingly, we're also just rolling out exclusive services and experiences that aren't available in the general marketplace such as airport lounges and access to select properties. So we continue to lean into growth here. And obviously, that quest towards the top of the market involves quite a bit of marketing investment and a lot of upfront investment for annuities that are just wonderful, long-lasting, fabulous annuities.
And the third vector of real marketing investment is our continuing efforts to build our national bank. And just as a reminder, we have a smaller branch footprint. And so we lean more heavily on our technology investments, our digital experiences, our cafe network, and our brand and marketing investments to continue, to organically build this national bank. And we are really pleased with the traction there, and it's been a lot of years in the making, but we are definitely leaning in there and love the results.
So Ryan, those are kind of the window into how we're thinking about it and the compelling opportunities behind -- across the board that we see and we are continuing to capitalize on the opportunity as we see it.
Got it. And Rich, if you put late fees aside for the first part of my question, you talked about stable to modest improvement, which escalate fees would imply continued improvement. On the operating efficiency, do you think we're back ex late fees on a sustained journey of improving efficiency, like you were talking about before the pandemic? And then second, just how are you thinking about the timing of offsetting late fees?
Yes. Thank you. So the -- our story about efficiency, I think, has been a very consistent story for a bunch of years. And it's -- I think a lot of companies drive efficiency by just continuing to cut costs on their way to greatness. And we are certainly -- put a lot of energy into the cost side of our business. But really, it's been about building a business model powered by our technology and the customer experiences we're building to drive revenue growth and efficiency, and -- both at the same time.
And we've talked about how so much of this is powered by technology, and we continue to see the benefits of that. So we -- at the -- on the one hand, keep leaning into technology and keep investing there, and on the other hand, see a growing opportunities to drive efficiency as a beneficiary of the technology investment.
So pulling way up, while any particular year can be different from overall trends. We continue to believe that an important part of the value proposition with investors, and the benefit of the years of investment we're making is to continue to drive greater operating efficiency.
So did you have a CFPB question? Yes. So let me turn, Ryan, to talk about that.
So the CFPB's late fee proposal as currently contemplated would reduce late fees by approximately 75%. While the CFPB's proposal has not been finalized, we expect the CFPB to publish a proposal soon, and once the CFPB publishes its final rule, we expect there to be industry litigation that could delay or block the implementation of the rule. This litigation will likely delay the implementation of the rule until at least the second half of this year and maybe longer, you saw, we talked about an estimate of October.
If the proposed rule is implemented, there will be a significant impact to our P&L in the near term, relative to what our path would have been. However, we have a set of mitigating actions that we're working through that we believe will gradually resolve this impact a couple of years after the rule goes into effect. And these choices include changes to our policies, our products and investment choices. Now some of these actions will take place before the rule change takes effect and a few are already underway, many will come after the rule change takes effect.
Our next question comes from the line of Arren Cyganovich with Citi.
I was hoping you'll talk a little bit about the auto business. You continue to kind of pull back a little bit from that side, thinking about how you're feeling about potentially increasing some originations, still a healthy amount of originations at $27 billion this year, but just wondering when you might make a pivot there.
Thanks, Arren. So we've been cautious in auto for a couple of years now. We've noted over this period of time, a number of headwinds in the business. So let's tally them up. Margin pressure from the interest rate cycle, normalizing credit vehicle values normalizing from their all-time highs and affordability pressures stemming from the combined effects of elevated interest rate and still high car prices. Now as you know, we don't work backwards from growth targets. And we remain disciplined in our originations, setting pricing and terms that we're comfortable with and then take what the market gives us. So back in 2022, we raised price, tightened our credit box at the low end of the market, and took other steps to manage the resilience of our lending. As a result, our run rate of originations has been lower than, like 2 years ago. But as a result of our actions, we've been just very pleased with the performance of our auto originations. The credit performance has been really striking, which, of course, you can see. So even as vehicle values continue to normalize, risk on our most recent originations from 2023 remains below what we saw in our pre-pandemic originations, probably is the result of our actions. And vintage over vintage, that risk remains stable. The margins on new originations have improved as well, particularly over the last couple of months as interest rates have come down from their recent peaks. So we feel quite good about the performance of our auto originations. So we continue to adjust our strategies where we see opportunities for growth or emerging risk. But of course, that's what we always do. But when we think about some of the headwinds, I think some of those headwinds are easing, and the results that we're seeing on our own book are really pretty striking and gratifying. So that gives us a more bullish outlook, still with a note of caution.
Our next question comes from the line of Rick Shane with JPMorgan.
Rich, you've given some framework for charge-offs into '24, one of the observations we would make is that delinquencies, even through December on a year-over-year basis, do -- are trending -- are up on a year-over-year basis even though that increase has slowed substantially. That suggests, very consistent with your description, that charge-offs will continue to rise through the first half of the year. What I'm curious about is given that delinquencies are still up 100 basis points, 115 basis points year-over-year in December, when we look into the second half of the year, I understand, seasonally, that they will be down. But would you expect the charge-offs in the second half of the year will still be up versus '23?
So I think the best way to think about this is to focus on the most stable benchmarks. Our focus here is about stability. So we're really looking at what are the most stable benchmarks that we can anchor to. And that really leads us back to 2019, 2018. And so -- let me just sort of speak in that, sort of double-click a little bit into my comments on that. So our delinquencies are above pre-pandemic levels, but they've been tracking with normal seasonality for quite some time. And now compared to 2019, since August, we're running around 17% above the level for the same month of 2019. Compared to 2018, since May, we're running at around 13% above the level for the same month of 2018. And at this point, we have a pretty good window into January of 2024 as well, based on delinquency entries in December, and that looks like it's going to be another month of stability. So we feel confident declaring that our delinquencies have stabilized. And of course, delinquencies are our best leading indicator of credit performance. Our charge-offs have been catching up to the stabilizing trend in our delinquencies over the second half of 2023. But at this point, what we're declaring here is that our charge-offs are leveling off as well. Now there's more month-to-month volatility in charge-offs than in delinquencies by quite a bit, because every data point of delinquencies includes 5 months of delinquency inventories.
And of course, charge-offs is looking at the relatively small number that falls off at the end of the last bucket. And there was also some noise in the fourth quarter of 2019 that makes it less reliable as a charge-off benchmark.
So we actually think 2018 is an even better benchmark for our charge-offs, for comparing our charge-offs in this fourth quarter that just happened, to a past stable year. In the fourth quarter, our net charge-offs were about 15% above 2018 levels. And of course, 2018 rolled into 2019. So that's an appropriate benchmark to look at as we head into 2024, and compare to 2019.
Now when we look ahead, extrapolating from our current delinquency inventories and recent flow rates, we conclude that our net charge-offs are stabilizing at about 15% above 2019, with, of course, some typical month-to-month volatility and normal seasonality. Now of course, the seeds of this stabilization have been planted for quite a long time now, and partly driven by the choices that we made back in 2020 and 2021.
Coming out of the pandemic, we were concerned about two trends. Fintechs were flooding the market, especially the subprime market, with credit offers, creating the potential for credit worsening and adverse selection in our originations. We also anticipated that pandemic era stimulus and forbearance would temporarily inflate consumer credit scores and that fees would revert over time.
So we tightened our underwriting in anticipation of these effects, and we have continued to make adjustments at the margin since then. And the result has been striking. That with all the kind of changes, the normalization, all the noise over the last number of years, that there has been strikingly stable performance on our origination vintages.
In our -- basically in our post-pandemic originations, each quarterly vintage for a given segment has been more or less on top of each other and also relatively consistent with pre-pandemic vintages. And over time, this just created a lot of stability that increasingly moved into our portfolio, and it contributed to the stabilization of our portfolio credit trends. And as we've looked at this, we said, these are very good shoulders to stand on, to have that much stabilization for so long. We, of course, all still waited to see exactly the manifestations, ultimately, of the portfolio stabilizing.
Another factor contributing to stabilization is our recovery rate, and unusually low recoveries have been the largest driver of our overall charge-off rate running above pre-pandemic levels. And this is, of course, because of the very low level of charge-offs over the past 3 years. And therefore, we had a low level of raw material for future recoveries. And by the way, just to Capital One point here. This is a larger effect for us than for most competitors because we tend to have meaningfully higher recovery rates than the industry average and because we tend to work our own recoveries, so they come in over time, not all at once, like in a debt sale. And so we've recently observed that our recovery rate has stabilized and started to tick back up -- and that's our -- and now that our recoveries inventory has started to rebuild, that's, of course, a good guy, although it's coming from a pretty low level. And this also contributes to our confidence that our overall loss trends have stabilized.
So when you -- to your question, where you compare to 2023, what we have really done is really kind of anchor our benchmarking to the most stable years in sort of recent experience, 2018 and 2019. And since we've seen delinquencies and charge-offs stabilize relative to like quarters in those benchmarks. We felt the best language with which to describe where things are settling out is to do it as a multiple of those two generally stable years. And so we are entering 2024 now with a real sense of stability. And we've benchmarked where we are as a multiple of those 2 stable year benchmarks.
Our next question comes from the line of Moshe Orbach with TD Cowen.
Great. Maybe, Rich, just following up on that. I think that a lot of the marketing dollars in the card space have been spent with on the transactor business. But given that stability that you're talking about, and I assume the bulk of the dollars of loss or delinquency and loss are coming from kind of the lower end of the card spectrum. So does that mean that that's an area for expansion in 2024? Like how should we think about that? And I do have a follow-up.
Sorry, Moshe, are you saying is what -- is the lower end an area for expansion.
Yes.
So we feel -- what we feel good about all of our segments across the credit spectrum in card and also the relative health of the consumer. And you've known us for a long time, Moshe, and as long as we have -- for the decades, we've been talking together. You know that we have a long history of delivering sustained resilience and profitability at the lower end of the marketplace.
And we -- let's just reflect on this for a second. If you're talking about subprime credit card, this is a complex business that requires deep investment in information-based underwriting. And of course, we've spent decades developing and testing tailored product structures and sort of honing the analytical and the operating and underwriting and marketing capabilities to attract and serve this franchise, but also with the #1 and 2 and 3 most important things to us has been resilient as we do this. And what's really been pretty striking is how consistent our strategy has been over the years through the Great Recession and following that. And if we -- and we've talked about how the lower end of the marketplace, whether you're talking -- by the way, there's a whole -- when you talk about the lower end of the marketplace, obviously, we there's a whole part of the marketplace that Capital One doesn't serve. But in terms of the lower end of the marketplace that we serve.
If you look at either income or FICO, and look at the normalization that's going on, we've seen very solid curing in that part of the market. It, in fact, even started -- it cured a little bit earlier than some of the other parts of the market. But the curing story, the leveling off story we're talking about today is absolutely across the board.
I do want to say, though, also Moshe, relative to your point, you've had fintechs who we were very concerned about flooding that into the market some years ago. They certainly have massively dialed back. And I think that the continued when we see the success of our vintages, the stabilization, now the sort of stabilization overall of Capital One's whole portfolio and the dynamics in the marketplace, I think that we like the opportunities we see there, Moshe, and we will be leaning into that.
Great. And just as a follow-up. I mean, you've talked in the past about not just the financial impact of the late fee but also its deterrent impact. So how do you think about that in terms of the resilience of that segment as we go forward post any changes to late fees?
And maybe just as a side point, you did mention that you thought you could improve the efficiency ratio even with the late fee. I mean, that just seems like you have to take a couple of hundred million dollars out of expenses to do that. So if there's a way to talk about what -- how that would happen, and tack that on to the answer that would help, too.
So Moshe, Capital One has pursued a strategy for the many years of trying to create and deliver to the marketplace, strikingly simple products. Because we -- some sort of a mission and strategy point of view, we believe so much in this, but -- and we built a brand over having very simple products. And -- so things like, for example, on the banking side, no minimum balance requirements, no membership fees and even no overdraft fees. So here, we are a company that has really, really reduced the fees, but if we had one fee left, I think the fee we would most hang on to is the late fee because, to your point, it plays a very important role in the deterrent value to a consumer. And an analogy that we sometimes use is speeding ticket. I think that if a speeding ticket were -- let's say, we had an $8 speeding ticket, I'm not sure that our highways would be quite as safe as they are now because if we're really trying to deter behavior that we think is really consequential for people, that really is the role of the fee. We've been very active in giving alerts to all of our customers.
When late payment do alerts with the goal of trying to -- not trying to maximize late fees, but actually trying to maximize the on payment performance of our customers. So Moshe, this is a question that we've been worried, about your question about the -- what could be the impact on credit performance of individuals. And it's something that we're just going to have to -- if this CFPB rule goes into effect, we're all going to experience together sort of this not controlled experiment, but we certainly mark us down for having a concern about that. But from a financial point of view, obviously, the late fees are an important thing on the P&L. And as I've talked about, we have created a set of many actions across different types of things from policies, products, pricing structures, investment choices to claw back the very significant economic impact. Some of those things are underway. Some of them -- just to mention it, by the way, by the time we get there, when the rule is announced, some of the offsets are going to by then be into the run rate of the company and a majority will still be waiting to happen. With respect to the fourth quarter, that '25 is the big full year effect. Obviously, something coming in late in the fourth -- something coming in, in our estimate, in the fourth quarter, doesn't have as much impact on the annual efficiency ratio, but it still does have an impact. So essentially, what's implied underneath it, is quite a bit of progress on the efficiency ratio behind the flat to modestly down guidance that includes that fourth quarter effect.
Our next question comes from John Pancari with Evercore ISI.
In the interest of time, given it's late in the call, I'll ask my two-parter all upfront here. First, on the marketing side, I know you indicated you expect to continue to lean in on marketing this year. Does that, what does that imply on how you're thinking about full year marketing expense? I mean, could that mean that you'll see marketing come in above the $4 billion level that you saw this year? Or could it be stable or a modest decline? And then my second question is on the credit side, on commercial real estate office, CRE. I know you had some lumpy losses this quarter and some pressure still in criticized in nonaccruals. If you could just give us a little bit more color there in terms of what you charged off and your outlook on that front.
Okay. John, thanks for your good questions here. We don't typically give full year marketing guidance. And the reason is because -- marketing depends, of course, a lot on the opportunities that we see when we get there. So what I wanted to just share in response to Ryan Nash's question is, a continuation in the positivity that we feel both about the real-time numbers we're seeing of response and performance of our vintages and all of that. And then also the sort of more structural investments that we're making in the business, particularly with respect to the -- going after the heavy spenders. So we don't have full year guidance, but we certainly continue to like the opportunities that we see.
And then, John, on the office side, it's virtually impossible to generalize office. It is incredibly property specific. We've talked in the past about us having a fair amount in gateway cities and having a mix of both A and B, C properties. But frankly, the decomposition matters a whole lot less than the individual properties. And so what we saw in the quarter was a little more than $80 million losses tied to office loans. We continue to not originate their balances have come down to about $2.3 billion, I think down about $150 million in the quarter, it's less than 1% of our total loans. But as we charged off in the quarter, we had essentially reserved entirely for that amount. And then we built back up a little bit for the remaining portfolio to maintain the coverage at around 13%.
Our next question comes from Don Fandetti with Wells Fargo.
Rich, you've made a lot of progress on heavy spenders. As you sort of look out, where are we, I guess, on that expense cycle? Are we -- are you sort of still looking at many years of acceleration? Or do you hit some type of level where there's some scale kicking in? Just trying to get a sense on where we are on that investment cycle?
Well, Don, it's certainly -- I think the quest to the heavy spender -- to win in the heavy spender in the marketplace, it will be a quest as far out as we can see in the same way it is for the players who -- the small number of players who are really going after that business. The key part of it is we're getting more and more scale along the way. So you've seen over the years, the growth in purchase volume, what you don't see is the purchase volume growth rates by a level of spender. And any segmentation we've been looking at, it monotonically -- the growth rate is monotonically faster. The more you go up towards the heavy spenders. So it's just indicating we're getting a lot of traction there. So I wouldn't want to say that we just have to do a blitz and then we're kind of done with the investment that the way that scale is achieved is by getting more and more customers in a business where all the players in the business, even including the largest, continue to invest in that business. But we're really pleased with the traction. And that's why we continue to invest.
Our next question comes from Bill Carcache with Wolfe Research.
Rich and Andrew, I appreciate all of the very clear commentary on what you're seeing in credit. There's a view that if we'd had a mild recession and experienced the purging of weaker credits that, that would have provided a clear runway for growth coming out of that. But instead, the environment we're in is arguably a little bit muddier and some would stay still late cycle. Could you speak to that dynamic, Rich, and whether that weighs on how you're thinking about growth from here in any way? And as a follow-up, I'll just ask it now for you, Andrew. Can you update us on how you're thinking about capital return from here?
Rich, why don't I start? Bill, look, at this point, there still remains a number of uncertainty around capital, not the least of which is the endgame proposal. We're all aware there's been quite a bit of advocacy there, and there remains a fair amount of uncertainty of where the rule will win, including things like the impact of AOCI and phase-in and ops risk and other forces at play. So we're -- you know as much as we do, and we're waiting to see what the final rule holds there. But in addition to that, we're coming up on CCAR, we don't yet have the scenario for this year. You look back at how impactful the scenario as well as the starting balance sheet is to those outcomes. You've seen our SCB fluctuate over the last 4 years from, I think, 10.1 down to 7. And now we're sitting here at 9.3. So waiting to get a little bit more clarity of what CCAR will hold. And then in addition, we continue to see a range of outcomes in our own growth projections. And finally, I'll just point to the economy, there's the consensus view is growing of a soft landing, but there's still quite a wide range of outcomes there. And so given all of those factors, we've chosen to operate for the last few quarters around 13%, we recognize that when we feel like we're in an excess capital position, that returning it is one way to create value. And under the SCB framework, we have that flexibility to manage repurchases dynamically, and we'll use that flexibility when we think it's prudent to do so.
Bill, a comment on our continuing to lean in, given that some people might argue that the economic environment is late cycle. So certainly a great question. So there -- first of all, the bottom line is we are continuing to lean in. Obviously, we keep a wary eye out for things that could change. But I sort of start with the health of the consumer. I think the U.S. consumer remains a source of strength in the overall economy. And the labor market has proven strikingly resilient over the past year, really defining the expectations of many economists in the face of rising interest rates. Consumer debt servicing burdens remain relatively low by historical standards. Again, despite rising interest rates, home prices are back and doing a bit better and they're generally near all-time highs. In aggregate, consumers across all income levels still have excess savings also from the pandemic, although those numbers are declining. Inflation has moderated to the point that real wages are growing again after shrinking for almost 2 years. Student loan repayments now -- they resumed in October, but there's the 12-month on-ramp period and a new income-driven repayment plan, which will significantly reduce payments for lower income borrowers. So on the whole, I'd say consumers are in really quite good shape relative to most historical benchmarks. Then if we look inside our own portfolio, we still see higher average payments compared to 2019 by segment, by a really, pretty sizable delta. We -- and then we -- we then look at the marketplace. And you've seen in the auto business, how at times we get alarmed by some of the practices or the pricing in the industry and we pull back in ways that we haven't pulled back in the card business. But I think we see a rational and stable competitive marketplace, it's very competitive, but it's rational and it's stable. And then most importantly, the results themselves. The -- our vintages just continue to come in on top of prior vintages, the trimming around the edges that we've done over the last few years have really, I think, allowed our results to have a stability to them that even has diverged from the sort of underlying not as good performance in the marketplace of things recently compared to the past, but we have that real stability. Then we see the leveling off of our portfolio. And really, we talk about our charge-offs leveling off at a level that's like 15% above 2019. It's interesting, actually, that's net charge-offs, but gross charge-offs are leveling off very close to the gross charge-off levels of 2018 and 2019. And actually, the thing that creates the differential is the lower recoveries that we've had for -- as a -- in the wake of the inventory recoveries being so much lower inventory of charge-off debt. So pulling way up and seeing the traction in our business, the success with our brand the things that, for competitive reasons, we don't share in the marketplace, but the traction on the tech side in terms of enabling us to create a better, really unique, customized customer experiences, totally customized underwriting, the reaching to marketing channels that we hadn't even tapped before, all of this is putting us in a position to continue to -- pulling way up, obviously, in the credit business, we always worry a lot. But if I calibrate this relative to a lot of other times, I feel really quite good about this. And I actually said I felt a lot less good, a couple of years ago because I felt that the pandemic, well, from a credit point of view, who couldn't like those credit results, I said it actually is so abnormally good. The marketplace won't be able to help itself but create unusual practices, unresilient underwriting, et cetera. So actually, what we've had, if I can borrow the soft landing term from the economy conversations is kind of a soft landing relative to the credit business, and landing is really quite the right word relative to Capital One, which I think really, as I've kind of declared today as sort of landed here. And I know some competitors still haven't fully landed. But pulling way up on this, I actually feel this is really quite a good time if I calibrate to all the times over the years in this exciting journey.
Our final question comes from the line of Dominick Gabriele with Oppenheimer.
I actually have two questions. Rich, what are you seeing, do you think that's making the net charge-offs stabilize 15% above 2019 levels? Is there something in the consumer payment behavior that's changed? Is there something you think that shifted the consumer in general, where the card industry may be seeing a higher through-the-cycle net charge-off rate going forward? Or for Capital One in particular. And I just have a follow-up.
So I actually believe that what we're really seeing here is a credit situation that's very similar to pre-pandemic. It is showing up right now. And I'm going to speak through the Capital One lens. I think, I'm not going to universalize for the industry. But as I mentioned in the -- to the prior answer there, that if you look at gross charge-offs, where they're settling out for Capital One. Now this is Capital One that has done a lot of trimming around the edges over the last -- a fair amount of trimming around the edges. I think we also did a very important choice that I'm not sure was universal. It might have been an unusual choice. But when we saw the incredibly strong credit performance of consumers, much of it driven by stimulus and forbearance, we sort of became alarmed about credit score grade inflation, if you will, and essentially intervened in our models to normalize so that we didn't get fooled by that. But it's -- this is the -- and so as a result of that, we have stabilized, we're probably one of the first players to stabilize, and we stabilized at this moment at 15% above, say, benchmark to 2019 levels. Already -- we said, that number from a gross charge-off number is really sort of very close to 2019 levels. So the recoveries effect, which is a temporary effect that our recoveries are so much lower because they just don't have as much inventory of charge-offs to collect on. That's a good guide that should help over time. So I think also as we have talked about, and we've talked about ever since the pandemic sort of -- we started coming out of the other side of the pandemic. We said there's another effect, let's call it, the delayed charge-off effect, that if you think about all those charge-offs that would have happened, many of that would have happened in the pandemic but didn't. Some of them may have gotten a reprieve for the long run, but a bunch of others we certainly have felt are going to charge off over time. And that is a temporary effect that we think has been playing out over this normalization thing. It's not -- we have ways that we try to measure it, but nobody can precisely measure this. But this is also something that leads to an elevation of charge-offs relative to probably what's an equilibrium. So if I speak from Capital One's point of view, our guidance was to guide you to the leveling off, because of the lower recoveries effect right now, that's leveling off at a 15% above 2019. The underlying credit dynamics seem very similar to me to what was there in the past. I think there -- even as we keep a wary eye on the economy, there's some just sort of actuarial good guys making their way through the business and all other things being equal, that can help the credit metrics more and more show that they are strikingly similar to what was there before the pandemic. So pulling way up, I don't think, and again, I'll speak -- I don't think things have shifted. I think we're seeing some trends playing out. But for Capital One, we're -- we feel great about where we've stabilized, and we see really a good assessment of our future.
Well, thank you very much, everyone, for joining us on the conference call tonight, and thank you for your continuing interest in Capital One. The Investor Relations team is available. So see me to answer further questions if you have them. Have a great evening.
This concludes today's conference call. Thank you for participating. You may now disconnect.