Capital One Financial Corp (COF) 2016 Q3 法說會逐字稿

Welcome to the Capital One third-quarter 2016 earnings conference call.

(Operator Instructions)

Thank you. I would now like to turn the call over to Mr. Jeff Norris, Senior Vice President of Global Finance. Sir, you may begin.

Thanks very much, Kevin, and welcome, everyone, to Capital One's third-quarter 2016 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 third-quarter 2016 results.

With me today are Mr. Richard Fairbank, Capital One's Chairman and Chief Executive Officer; Mr. Steve Crawford, Capital One's Head of Finance and Corporate Development; and Mr. Scott Blackley, Capital One's Chief Financial Officer. Rich and Scott will walk you through this presentation. To access a copy of the presentation and 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. For more information on these factors, please see the section titled forward-looking information in the earnings release presentation, and the Risk Factor section in our Annual and Quarterly Reports, which are accessible at the Capital One website and filed with the SEC. And with that, I'll turn the call over to Mr. Fairbank. Rich?

Thanks, Jeff, and good evening, everyone. I will begin tonight on slide 4, with our Domestic Card business. Loan growth and purchase volume growth remained strong. Compared to the third quarter of last year, our ending loans grew $8.8 billion, or about 11%. Average loans were up $9.4 billion, or about 12%. Third-quarter purchase volume increased about 12% from the prior year. We continue to like the return and resilience profile of the business we are booking.

In the quarter, we also announced that we entered into an agreement for a new partnership with Cabela's, and the acquisition of Cabela's co-branded card portfolio, which has roughly $5.2 billion in outstandings. The agreement is subject to regulatory approvals and customary closing conditions. Cabela's is a great retailer with a powerful brand and highly engaged and loyal customers.

Revenue for the quarter increased 8% from the prior-year quarter, slightly lagging average loan growth. Even with the positive margin impact of higher subprime mix, revenue margin declined year over year, as expected with our exit of the back-book of payment protection products at the end of the first quarter. Revenue margin for the third quarter was 16.6%. Non-interest expense increased 4% compared to the prior-year quarter.

Our Domestic Card business continues to gain scale and improved efficiency. Net interchange revenue for the total Company grew 9% from the prior-year quarter versus the 12% growth in Domestic Card purchase volume.

As we have discussed, there is considerable quarterly volatility in the relationship between these two metrics. For the past several years, on an annual basis, net interchange growth has been well below Domestic Card purchase-volume growth. We'd expect this difference to continue as we originate new rewards customers in our flagship products and extend rewards to existing customers. Additionally, a few of the largest merchants have negotiated custom deals with the card networks. These deals are putting pressure on interchange revenue, and we expect the pressure to continue.

As we've discussed for several quarters, the dominant driver of year-over-year charge-off rate trends is growth math. Which is the upward pressure on delinquencies and charge-offs as new loan balances in our front books season and become a larger proportion of our overall portfolio, relative to the older and highly seasoned back-book. Growth math began to impact charge-off rates in 2015. We still expect the peak impact of growth math in terms of its contribution to year-over-year change in our loss rate to be in 2016, with a diminishing effect in 2017, and only a modest effect beyond that. In the third quarter, growth math drove the increase in charge-off rate compared to the prior year, and seasonality drove the improvement in charge-off rates compared to the linked quarter.

Looking ahead, two effects are together moderately impacting our charge-off outlook since last quarter. One effect is better-than-expected growth, especially in subprime. While we are growing nicely everywhere we are investing, we are seeing particular success in subprime, which has raised the mix modestly. Our proportion of loans below FICO 660 has grown from 35% one quarter ago to 36% at the end of the third quarter. Subprime has higher losses than average, and they are also more front-loaded, so it tends to have a pretty immediate impact on our near-term credit metric.

Beyond the growth and mix effect, we have revised slightly upward our front-book loss expectations for 2017, and slightly downward our front-book loss expectations for 2018, based on the composite performance and projections for hundreds of credit programs. These effects together lead us to raise our full-year 2017 charge-off guidance from the low-4%s to the mid-4%s, with normal seasonal variability and excluding the modest benefit we expect from adding the Cabela's portfolio. With nine months of actual results in the books, we expect 2016 full-year charge-off rate will be around 4.15%.

We expect the changes in guidance to largely play out over the next three or four quarters, so the expected impact of higher charge-off is mostly captured in our current allowance. While we're modestly raising our loss outlook for 2017, our internal view of 2018 losses is unchanged, with slightly improved expectations for front-book performance offsetting higher subprime mix.

Pulling up the higher growth in subprime mix are also driving up our revenues and pre-provision earnings. We continue to see attractive growth opportunities in our Domestic Card business. But it's clear this opportunity won't last forever. The marketplace is moving, and competitive intensity across the card business remains high, and revolving credit growth is now 6% year over year. We will continue to monitor the marketplace vigilantly. But in the meantime, we continue to like the opportunities that we see.

Slide 5 summarizes third-quarter results for our Consumer Banking business. Ending loans grew about 2% compared to the prior year. Growth in auto loans was partially offset by planned mortgage runoff. Ending deposits were up about $8 billion or 5% versus the prior year. Third-quarter auto originations were $6.8 billion, about 22% higher compared to the third quarter of last year, with strong growth in prime, near prime and subprime.

Similar to our Domestic Card growth, we like the earnings profile and resilience of the auto business we are booking, and continue to believe that the through-the-cycle economics of our auto business are attractive. The auto market and competitor practices remain dynamic. While we see opportunities for growth, we remain very vigilant about competitor practices. Our underwriting assumes a decline in used car prices. We continue to focus on resilient originations, and we continue to expect a gradual decrease in margins and a gradual increase in charge-offs as the cycle plays out.

Consumer Banking revenue for the quarter increased about 3% from the third quarter of last year. Higher revenue from growth in auto loans and higher deposit volumes was partially offset by margin compression in auto and planned runoff of mortgage balances. Non-interest expense for the quarter also increased 3% compared to the prior-year quarter, driven by growth in auto loans and an increase in retail deposit marketing.

As we've discussed in prior quarters, we've been optimizing both the format and number of branches to better meet the evolving needs of our customers as banking goes digital. In the third quarter, actual changes related to branch moves were about $60 million. Year to date, we've recognized about $106 million of the $160 million in expected costs for 2016. We expect branch optimization costs to continue in 2017. These costs show up in the other category, rather than in the Consumer Banking segment.

Third-quarter provision for credit losses was up from the prior year, primarily as a result of charge-offs, in addition to the allowance for loan losses for the auto portfolio. Growth in auto loans, the expectation of gradually rising auto charge-off rates, and the expectation of declining used vehicle values, drove the trend in consumer bank provision for credit losses.

For several quarters, we've said that we expect pressure on our Consumer Banking financial results. In the home loans business, planned mortgage runoff continues. In auto finance, margins are decreasing and charge-offs are rising modestly. And our deposit business has continued to face a prolonged period of low interest rates. We expect that these factors will negatively affect Consumer Banking revenues, efficiency ratio and net income, even as we continue to tightly manage costs.

Moving to slide 6, I'll discuss our Commercial Banking business. Third-quarter ending loan balances increased 28% year over year, including the acquisition of the GE Healthcare finance business. Excluding the $8.3 billion of loans acquired from GE, ending loans grew about 12% over the same time period. Average loans increased 28% year over year, while average deposits increased 2%. Revenue was up 27% from the third quarter of 2015.

Credit pressures continued to be focused in the oil and gas and Taxi Medallion portfolios. We've provided summaries of loans, exposures, reserves and other metrics for these portfolios on slides 14 and 15. For the total Commercial Banking business, third-quarter charge-offs were $108 million, primarily driven by charge-offs of Chicago Taxi Medallion loans, and to a lesser extent, oil and gas loans.

The charge-off rate for the quarter was 66 basis points. We added to reserves for some third-quarter weakness in New York City Taxi Medallion values, but the reserve additions only partially offset the charge-offs flowing out of reserves, resulting in a net reserve release in the quarter. Combining both charge-off and reserve changes, provision for credit losses declined $14 million from the prior-year quarter to $61 million.

Criticized and non-performing loan rates were relatively stable in the quarter. The criticized performing loan rate for the quarter was 3.7%, and the criticized non-performing loan rate was 1.5%. Now I'll pass the call over to Scott.

Thanks, Rich. I'll begin on slide 7. Capital One earned $1 billion or $1.90 a share in the third quarter. Excluding adjusting items, earnings per share was $2.03. Adjusting items in the quarter included a $63 million build in our UK Payment Protection Insurance customer refund reserve, of which $47 million was an offset to revenue and $16 million was captured in operating expense. A slide outlining adjusting items can be found on page 12 of the slide deck.

Pre-provision earnings increased modestly on a linked-quarter basis, as higher revenues were only partially offset by higher non-interest expenses. As I highlighted last quarter, we had a one-time rewards expense that reduced net interchange income by $38 million, driven by the completion of some system enhancements that moved our rewards liability cutoff to the last day of the quarter. Provision for credit losses was flat on a linked-quarter basis, as higher charge-offs were almost entirely offset by a lower-linked quarter allowance build. An allowance roll-forward by segment can be found on table 8 of our earnings supplement.

Let me take a moment to explain the movements in our allowance across businesses in the quarter. In our Domestic Card business, we built $349 million of allowance. This build was driven by growth and subprime mix in the quarter, and incorporating our charge-off expectations for the next 12 months into our allowance calculation. In our Consumer Banking segment, allowance increased $31 million in the quarter, almost entirely driven by growth in our auto business.

In our Commercial Banking segment, we had a $48 million reduction in reserves, driven by charge-offs in our Taxi Medallion portfolio. Recent reserve movements have focused on Taxi and our oil and gas portfolios, and we have provided details for those portfolios on slides 14 and 15 of the appendix of tonight's slide deck.

Turning to slide 9, you can see that reported net interest margin increased 6 basis points from the second quarter to 6.79%. That was primarily driven by day count. NIM [MIG] also increased 6 basis points year over year, which was fueled by strong growth in our Domestic Card business.

Turning to slide 10, I'll discuss capital. As previously announced following the approval of our 2016 CCAR capital plan, our Board authorized repurchases of up to $2.5 billion of common stock through the end of the second quarter of 2017. In the quarter, we accelerated the pace of our buybacks and repurchased 17 million shares or $1.2 billion of our authorization. Our common equity Tier 1 capital ratio on a Basel III Standardized basis was 10.6%, which reflects current phase-ins. On a standardized fully phased-in basis, it was 10.5%. And with that, I'll turn the call back over to Rich.

I'll close tonight with some thoughts on third-quarter results and our outlook as we head into 2017. We posted another strong quarter of growth in Domestic Card loan balances and purchase volumes, as well as growth in auto and commercial loans, driving strong year-over-year growth in revenue and related increases in operating expense and allowance for loan losses. We've been working hard to improve efficiency by growing revenues, realizing analog cost savings and other efficiency gains as we become a more digital Company, and tightly managing costs across the enterprise.

Our efforts are paying off. Our efficiency ratio for the quarter was 52%. Excluding the impact of UK PPI, the adjusted efficiency ratio was 51.4% for the third quarter and 51.8% year to date. It's clear that we are on track to deliver much more than our prior guidance of some improvement in our full-year 2016 efficiency ratio, excluding adjusting items. Even with the expected significant seasonal increase in non-interest expenses in the fourth quarter, we now expect that our full-year 2016 efficiency ratio will be substantially lower than full-year 2015.

In 2016, we're already on pace to deliver the efficiency ratio improvement we had expected to achieve in 2017, primarily driven by enhanced revenues and concerted efforts to drive out analog costs. From this more efficient starting point, we expect that our near-term annual-efficiency ratio, excluding adjusting items and the expected impact of Cabela's, will be in the 52%s, plus or minus a reasonable margin of volatility. Over the longer term, we continue to believe that we should be able to achieve gradual efficiency improvement, driven by growth and digital productivity gain.

Pulling up our strong growth over the last two years puts us in a position to deliver solid EPS growth in 2017, assuming no substantial change in the broader credit and economic cycles. We expect that revenue will grow, and will drive growth in pre-provision earnings as well. We expect the upward pressure from growth math on the allowance for loan losses will begin to moderate. And we are reducing share count.

We continue to be in a strong position to deliver attractive shareholder return, driven by growth and sustained returns at the higher end of banks, as well as significant capital distribution, subject to regulatory approval. And now Scott, Steve and I will be happy to answer your questions.

Thank you, Rich. We will now start the Q&A session. As a courtesy to other investors and analysts who wish to ask a question, please limit yourself, as usual, to one question, plus a single follow-up. If you have any follow-up questions after the Q&A session, the Investor Relations Team will be available after the call. Kevin, please start the Q&A.

(Operator Instructions)

Ryan Nash, Goldman Sachs.

Good evening, guys. Rich, thanks for the update on the credit guidance. Can you maybe just flush out a couple of those comments for us? And in particular, what it means for the allowance? You talked about higher charge-offs, but that largely being in your allowance already. So I guess just, one, what is driving the changes in the back-book expectations for 2017 and 2018, and is that just mix?

And you noted that you still think that you're on track to deliver solid 2017 EPS growth. You referred to it as a coiled spring. Do you still think that's achievable, given your expectations for slightly higher charge-offs?

Okay, Ryan, there's a bunch of great questions that you have. I think, Scott, you can take the allowance thing. So first of all, when you said to tell me about your change in back-book expectations, we don't really have any change in back-book expectations. The change in expectations that we are talking about is driven by two factors. And a key part of this point is that better-than-expected growth, especially in subprime, is a key part of our revised guidance. And as for our front-book loss expectations, we revised them up.

Yes, sorry, Rich. I meant front book.

Yes, okay, exactly. So as the front-book loss expectations, in other words, beyond the effect of the growth and mix, so I'm trying to de-average all these effects, there's a slight upward revision that we made in the quarter for 2017, and a slight downward revision in 2018. And that's driven by a compositive, like many multiple vintages across hundreds of lending programs.

And just giving you just a little bit more granularity behind this, if you compile them all and pull way up for the patterns that we see, 2015 programs are coming in with slightly higher losses than 2014 programs. And now, 2016 programs are now coming in a little bit better than 2015 programs. And so the higher 2015 origination losses make their way into the late 2016 and 2017 numbers. The lower 2016 origination losses make their way into the later 2017 and 2018 numbers.

Now, these are all very modest effects, is the most important point I want to leave with you. The other thing is, our view of the value creation from our lending programs is not only bullish and high. Just to put it in perspective, this thing we're calling the front book, which is basically everything that -- all the growth we've had since the higher growth began. So we're talking about 2014, 2015, 2016. Our view of the value creation for that is at the high end of anything we booked in the last 15 years.

So back to your question about how we feel about the earnings potential in the business. What you actually have is a card business that you kind of pull up on a yearly basis, that has maintained pretty flattish earnings, strong earnings power, while absorbing a whopping increase in especially front-loaded credit costs that come in the form of originations. That are, A, higher than the back book; B, that have their losses more front-loaded; and C, of course, it flows right through allowance.

And so this brings a lot of pressure right up front when we grow at the kind of trajectory that we are talking about. But hopefully you can see when you look at the numbers and do the math on what we are booking here, this is a coiled spring of a lot of earnings power. And that's why we continue to work hard to originate and grow what we can very prudently during this window of opportunity. Scott, why don't you comment about the allowance part of the question?

Ryan, let me spend a minute and just walk through some of the puts and takes in the allowance. One way to look at the build is to start off by taking the coverage ratio from last quarter and applying it to the change in ending balances. That really ignores the effects of growth and subprime mix that are impacting the loss rate. But you can see that if you start with that static coverage ratio and apply it to the change in balances, that's going to explain roughly a third of the build. Be mindful that when you get to Q4, if you're going to try to do that calculation, you need to adjust for seasonal balances. So just keep that in mind.

Now, the rest of the build is really associated with the loss rate and the mechanics of doing the calculation of the allowance. Rich just walked through in detail the drivers of our updated charge-off guidance growth in mix and performance. And so most of that guidance change happened during periods that are included in our Q3 allowance. And so we captured those as part of our updated allowance calculation this quarter.

In addition to that, the mechanical effect of the quarter swap, which is moving forward the lost forecast one quarter, results in us dropping off Q3 2016 and adding Q3 2017 to the allowance calculation. Our forecast, as you know, and as we've been saying, is for losses to be higher in 2017 versus 2016. So that change in quarters, all things equal, adds to the allowance calculation. And on top of that, we always have qualitative factors, which are for risks that are outside of the models.

So just pulling up though, looking ahead, I want to give you some of the building blocks as you think about how you can calculate the estimates of the allowance movements in 2017. So we told you a couple of key things. One, we told you what the full-year loss rate for 2017 is expected to be. We told you that the increase in our loss guidance has already been captured in the current allowance.

And we've told you that we expect the impact of growth math is diminishing in 2017, and only has a modest effect beyond that. So all those factors, I think, collectively give you a good base to start building up what your loss expectation is going to be in your 2017 allowance.

Thanks for taking my question. I'll step aside so others can ask questions.

Thanks, Ryan. Next question, please?

Eric Wasserstrom, Guggenheim Securities.

Great, thank you for taking my question, and thanks for that very detailed explanation about the ALL outlook. If I could just follow up, Rich, on the commentary about the coiled spring, which Ryan also referred to, should we expect 2017 net income, given the factors that you've just highlighted, to be above the 2016? And I guess the issue I'm really getting to is, we're seeing this outstanding growth at incredible marketing efficiency. And yet, we're seeing ROE decline, despite the impacts of the share repurchase. So I'm just trying to understand what we're looking for in terms of secular profitability?

Eric, I don't want to go beyond the forward-looking statements that we made in the prepared remarks. But I think the key way to look at this, again, is, a pretty dramatic growth for several years, on the order of what we are talking about, puts a lot of pressure on the P&L of the card business. And it has been absorbing that. Now meanwhile, and I always use this term of a coiled spring, I think it's a great metaphor, but meanwhile, when we look at the value of what we are originating here, we're very bullish. And that's why we are growing at the rate that we are.

I think that all I would say is, again, just think through the mechanics of the timing of how this plays out. You start with a lot of growth, which has more costs, but especially, most importantly, impacts the credit metrics, and especially goes right through the allowance line. So how the earnings power shows up, the line item where the earnings power will really start to hit that inflection point, is really as allowance. Allowance builds start to mitigate. And that, of course, happens right at the time when, looking ahead, the growth math effects are also starting to really settle out.

And our point is, we are approaching that period in this dramatic period of growth in the card business, and I think that, that's why we're not giving specific guidance on earnings power. We're just saying that we should start seeing earnings power play out over time as the growth math does its very predictable thing.

Great, thanks, that's very helpful. Thanks very much.

Next question, please.

Don Fandetti, Citi.

Thank you, Rich or Steve, I guess I wanted to clarify. It sounds like you said that the 2016 vintage is coming in better than the 2015, and 2015 worse than 2014, which makes sense. It just seems -- can you talk a little bit about how 2016 could possibly be coming in better, and why that might be the case?

Right. So we look at -- there's many ways we look at that. But starting with mechanically, the actual vintages themselves and the very early credit metrics, things like delinquencies on those vintages. Additionally, Don, we also just look at the mix of what are the metrics, things like credit scores and so on, of the business that we are booking., what is that blended distribution. All of the key indicators we look at, and this matches our own intuition about what's going on, as well, show a positive impact for 2016 versus 2015.

Now I want to make it very clear, 2015 is going to be a very successful year and very profitable, and with very good long-term credit performance. We are talking about small changes relative to expectations as we revise, as we always do every quarter, our own credit outlook. We're just painstakingly trying to break down the components so you get a better understanding of this. But we feel great about 2014 front book, we feel great about 2015 front book. 2016, all the early indicators of 2016 suggest that from a credit point of view, it's going to come in a little bit better than 2015.

That's real helpful. And just quickly, you've been pretty active in subprime cards. Do you expect the other banks to reach down? You saw JPMorgan talk a little bit about it as early as Q2. Can you talk about competition in that segment?

Yes, it's a striking thing. We've talked a lot about industry growth, and we talk often about industry growth has gone from very low single digits, or even occasionally negative a couple of years ago, to 6%. And that has our attention.

But here's another stat that is an interesting one. If you look at the composition of the industry's growth over the last 12 months, and I'm going just to exclude Capital One from the whole calculation, because your question is an industry effect there, about 31% of the growth is in subprime. And so subprime growth has certainly picked up now. It's growing faster than prime, and certainly that has our attention. Now I do want to say, on the other hand, it remains well below pre-recession levels in absolute terms. Because there was such an exodus from that part of the business over the years following the Great Recession.

So at times, I think you hear from some players: subprime, we don't do that. But all I'm saying is, 31% of all the growth is subprime, and somebody is doing it. So that has our attention, yes, and it's an important part of the overall -- if you look at the industry story. And this, by the way, is the number one thing that we worry about, more so than the next recession. Obviously we worry about things like that too. But the industry growth being at 6%, and the subprime component of that, and then also beyond card growth, when you look at student lending, auto lending, installment lending, that has been running for a while now at around 7%.

So we are pretty obsessive about these things. It causes us to believe, first of all, that we have a window, and it's just another reason we've doubled down and gone really hard while we have this window. It means we have to be doubly and triply obsessive about looking for indicators of things that inevitably happen when supply goes up a lot. And those indicators can be broad-based impact on credit metrics, as well as, of course, just directly affecting response rates and things like that.

So we are on the lookout, and as we always do for our investors, try to raise these flags. But we feel very good about what we are booking. We like the resilience of what we are booking, and we still are moving forward with resolve to continue to capitalize on these opportunities. But there is no doubt the industry growth data is just one more reason to remind ourselves that everything is a window of opportunity, and we have to carefully watch the industry from here.

Next question, please.

Betsy Graseck, Morgan Stanley.

Hi, good morning. Or good afternoon, it feels like morning here. I'm just wondering on Cabela's, maybe if you could give us a little bit of some color around how you expect it's going to be impacting various line items in the business overall? Any investment spend that you need to do to prepare for it? And if there is any of what you expect Cabela's to do your numbers in the commentary that you gave on the outlook for 2017?

Hey, Betsy, this is Scott. In terms of our guidance -- so all of the guidance that we gave is excluding Cabela's. We haven't yet finalized all of our projections in terms of impacts for Cabela's. We think it's a fantastic partner. We're extremely excited to have won that business. It's a powerful brand, incredibly loyal and engaged customer. So we are pretty thrilled about that. We're still working on all the details of the closing of that transaction, so don't have more information about that at this point.

Okay. And then just separately, follow up on the vintage discussion that we just had. Just wondering, I know you mentioned that 2016 is seasoning better than 2015. Are we talking about the first quarter of 2016? Just wanted to get a sense as to how long you think it takes for things to season, and why the trajectory so soon into the 2016 cycle is looking better?

Well, you know, this is all something that, with every passing month, gives you a little bit more clarity. And that's why on all of this stuff, we triangulate from just the literal vintage analysis, but we already looking at all the vintages of 2016, can see a gap relative to early vintage reads versus 2015. In the same way that 2015 had a gap relative to 2014.

But again, it is a triangulation, or its reinforced by just kind of watching the underlying dynamics of even the mix within the mix of all the programs, and looking at the distributions of who is applying. It's something that's composite across hundreds of programs. But all the signs are consistent with us, that there is this -- you know, all of this we're talking about is modest effects, but enough that it was worth mentioning that 2016 is a positive, relative to 2015.

Next question, please.

Arren Cyganovich, D.A. Davidson.

Thanks. I guess relative to the window of opportunity do you still see in Domestic Card and thinking about how much that the subprime has grown, do you still expect to see that higher take-up rate on the subprime? Or do you have any kind of expectations for limiting that at some point if it grows too large relative to the other piece of the portfolio?

Well, Arren, one thing we have kind of been featuring subprime here. I don't want to diminish the really good momentum that we have in pretty much all of the parts of the card business that we are investing in. So we are pretty bullish about our growth opportunity in all of the areas. We've had more recently kind of a surge on the subprime side, but who knows. I don't want to project what will happen to mix on a relative basis.

I think we like all the things we are booking. I think we have reasons to believe we can continue to grow in all of the areas. But it was worth highlighting, though, the kind of recent surge that we've had on the subprime side. But I wouldn't extrapolate to that to some significant continuation of that surge.

Okay, thanks. And then just secondly, wondering if you had any update on the Capital One 360 Cafes that you have been testing in different markets, if it's too soon for anything you can share from the impacts of putting those stores into various cities?

The Capital One Cafes, they are to pull way up. As you know, we are a national Company in just about everything we do, except we are local in really only 20% of the nation in retail banking. So Capital One Cafes are the manifestation of how we are re-imagining the future of banking. And also how we can enhance the national banking presence we already have through our old ING book, that of course we call Capital One 360 now.

But the key thing there is, it's not about putting a branch on every quarter. Far from it. It is about building kind of, if you will, flagship stores in key metropolitan areas, and using that as, in many ways, as a manifestation of banking re-imagined, relative to the very traditional way that banking has been done for forever, basically.

We have been gradually expanding our footprint in the cafes. We've been very pleased with the early results from the cafes. And what we see is reinforces our view that this is a good thing to continue to build out on a very thin basis, distribution in major metropolitan areas, with these cafes.

Thanks, Arren. Next question, please.

Matt Burnell, Wells Fargo Securities.

Good evening, thanks for taking my question. I appreciate all the detail you've provided on the credit card side of things in terms of your expectation for losses. The question I frequently get from investors is your outlook for the auto portfolio, given that, that's also growing at a relatively healthy pace. Can you give us a sense as to and maybe a similar way of thinking about losses in originations in 2016 versus 2015? And what your thoughts are in terms of forward-looking losses for the auto portfolio?

So thanks, Matt. I'm not going to give you specific numerical guidance about the auto business. But let me just pull way up about auto. As we've talked about, first of all, we really like the auto business. I think it plays to Capital One's strength. And we built a very strong and kind of leading presence in that space.

The auto business after the Great Recession was something that I don't think we will see again in our lifetimes, in terms of unique opportunity to grow and create value, with high margins and exceptional credit quality. Because so many players had headed for the hills, and consumers were, gosh, they were even at times, walking away from their houses and still paying on their cars. And used car prices were just at incredible highs.

So much of our commentary about auto has been predicting, saying: it's got to be that this thing will normalize over time. And the journey since then has really been one of normalization, where pretty much every year, the margins are going down, every year credit losses have kind of gone up.

Now in Capital One's case, two things I want to say. We've also had a mix change going on within Capital One over the last few years that has been more market growth. And so that has tended to dampen this effect. But let me also say, really, if you stacked vintage after vintage, year after year, in auto, they have been, the last few years, going up year over year. But the overall thing that we have seen is, we have been -- surprised is probably an overstatement. But I think we have kind of predicted something, a more rapid normalization than we have actually seen.

All of that said, I think the auto business is still in a good place for us to generate profitable business. But it is mostly kind of normalized, and it has a number of things going on in the industry that cause us to be very vigilant, most importantly, kind of underwriting practices there. But I would say really over the last few years, the actual credit performance has been strikingly good, and maybe even a little better than expected.

Next question, please. Oh, I'm sorry, Matt. Do you have a follow-up?

Yes, just to follow up, if I can, for Scott, on the buyback. If I take the average price of Capital One shares over the course of the third quarter, that was about $68.68 versus the 17 million shares you said you re-purchased. Is a little bit less than $1.2 billion in buybacks versus the $2.5 billion announced buyback program for the four quarters of the capital planning cycle. Can you give us a sense, Scott, as to how you all are thinking about the buybacks for the next three quarters?

You can see that this wouldn't be a ratable use of the buyback. So I think that's the one thing that we'd point out. And obviously we're trying to be opportunistic, and look at the market and use the authorization that we have within CCAR in the way that we feel is most productive. And that's really what has led to the acceleration year to date in the total amount of repurchases we have authorized for the year.

Thanks, Matt. Next -- oh, I'm sorry. Go ahead.

Just to finish that thought, so looking forward, Steve, should we think about acceleration relative to a ratable buyback over the next three quarters?

Well, I don't -- again, we don't have a tremendous amount of flexibility as it is, with our repurchase authorization, so I don't think I'm going to get more specific. Obviously we have the limitation in CCAR that we can only do the amount authorized for the year. And it's really just a function of looking at the marketplace and where we are and how quickly we want to use that authorization. But I think we've demonstrated that you shouldn't count on it always being ratable going forward.

Thanks, Matt. Next question, please?

Chris Donat, Sandler O'Neil.

Good afternoon. Thanks for taking my question. Wanted to first ask on the efficiency of your marketing. We saw marketing spend go down 6% quarter on quarter. I know it will bounce around a bit, but with your Domestic Card growth in the double digits it looks like you're being very efficient with that spend. I'm just wondering if you cracked the code to higher efficiency, or there's something temporary going on with the lower spending on marketing?

On marketing, I wouldn't put any particular emphasis on what you see in the third quarter. As you know, the fourth quarter is seasonally, by quite a bit, our highest quarter. And there are reasons for that, of course. And turn on the television, you probably can see that, because we are sponsoring bowl games, and the ESPN Bowl Week. And there's a lot of things that we line up seasonally in terms of our marketing blitz in the fourth quarter.

The other thing I want to say about marketing is that, while I think on a pulling way up, calibrated across the years, I think marketing efficiency is at a pretty high level right now. It's maybe a commentary on our marketing, but probably as much a commentary about the window of opportunity we have to grow and the success of that. One thing is very clear. Competitors have stepped up their marketing. Competitors are stepping up their advertising. Competitors are working in a number of cases have, in fact, sort of duplicated some of the Capital One products themselves.

This game will require a pretty high level of marketing. And that's probably the case even as the growth opportunity wanes a little bit, just because it's the nature of how this competition works. So we feel great about our marketing, but to continue to succeed is going to take a fair amount of marketing.

And just shifting to the oil and gas, your held-for-investment portfolio decreased quarter on quarter, but unfunded exposure increased only by $100 million or so, and mostly in midstream. Is that just a reflection of confidence in that portion of the oil and gas portfolio?

This is Scott. I think when you look at that portfolio, we obviously see some opportunities in midstream. We think that there's been a lot of rationalization that has happened in the E&P sector as well. And then in oil field services, that's probably the area where we continue to see risk. Those are companies that are still struggling with the effects of reduction in capital spending. So that is an area where, in midstream, we are seeing some selected opportunities, and taking advantage of those.

Thanks, Chris. Next question, please.

Sanjay Sakhrani, KBW.

Thank you. I guess I have another question on the loan loss reserving from here on out. Scott, maybe you can go through it a little bit more. I want to paraphrase a little bit what you said. It seems like you probably have one more quarter of pressure in terms of build, because you are lapping a tougher quarter and taking on an easier one. But thereafter, you're going to grow the provision in line with low growth. Is that a fair assumption to make?

Thanks for the question. And just to go back, what we have told you is that we expect losses are going to grow into 2017. We haven't given you any kind of specific quarterly guidance, so I'm not going to go into how you might calculate the quarterly swap. But I do think you've got the right thread, which is that we do expect the impact of growth math to be diminishing in 2017, and with a modest effect beyond that. That's going to meaningfully impact the way the allowance works out as we get into 2017.

Okay. And I guess a follow-up question for Rich, just on Cabela's. You've talked about how these large deals are pretty aggressively priced historically. What struck you with Cabela's that made it quite worthwhile for you to go for it?

Well, I've been pretty vocal about the nature of the auction environment and how easy it is for companies to get caught up in the frenzy of this kind of bidding. What we have said is, on a selective and disciplined basis, we really do want to grow the partnership business. We want to grow this business where we can find partners that have attractive, loyal and growing customer bases, compelling value proposition and who, their view of what they are trying to achieve through a partnership, is in fact very consistent with what we would want to do, and very customer focused.

Now, Cabela's is an amazing franchise. I think anybody who walks into one of these Cabela's sees this is not your typical store. This is all designed around the customer experience. Customers spend -- the amount of time the average customer spends in a Cabela's is actually measured in terms of hours. It's different from your typical kind of shopping experience. So we're attracted to the customer focus that they have, and it's all about brand and franchise and that kind of thing.

So what we did is, enthusiastically went after this, but said that it still has to be within the context of a financial deal that really works for us, and of course, that works for Cabela's. In this particular case, I think that we found a meeting of the minds, and I think Cabela's liked the opportunity they saw with us. And we were able to maintain our financial discipline and actually win this transaction.

Thanks, Sanjay. Next question, please.

Chris Brendler, Stifel.

Thanks, good afternoon. Actually, I have a follow up to the last question. Just overall, your opinion of the private label market, the attractiveness of that market, the competitiveness. It seems like these deals are difficult to win. Can you talk at all of how many merchants or retailers you have added to the portfolio since you bought it from HSBC?

It seems that this is sort of a unique situation here, with Cabela's being rather public. But have you had success winning other retailers? And do you like this business? Any disclosure on how fast it's growing relative to the rest of the card business would also be helpful.

And then a follow-up question on a different question on the share count is down a little over 3% this quarter. Nice reduction. Is that all buyback, or is there anything else going on there? Thank you.

Okay. So first of all, your question was about private label. One thing I want to say is that the Cabela's deal is actually a co-brand partnership. So let me just pull way up there. In what we call our partnership business, there are two types of credit card programs. One is private label, which is not a Visa-MasterCard. That is just a retailers card. And the other is co-brand, where the retailer or the issuer is issuing a Visa card or a MasterCard, and there is spend both at the store and outside of the store.

As it turns out, the key to a co-brand program and a really good co-brand program, is kind of the Holy Grail. That's what -- you probably talk to any card issuer, I think the great -- a couple of handfuls, there may be a little more than that, of like truly great co-brand programs.

And the thing that characterizes it, in addition to strong retailers and that kind of thing, is where you see high out-of-store spend to supplement good in-store spend. And when you see that, then all the strong economics and the way credit cards work is a great thing. And that's why there's a lot of -- all the card issuers are really trying to get those great flagship programs. I would absolutely put Cabela's in that category.

Now, private label, again, it's a different business and not a co-brand business, by definition. All the spend that's going on is inside the store. And there, I think our focus has been going after the really top-of-the-line retailers, and we have not been going after the middle or lower end of that marketplace.

So here you see us where we've had a lot of growth and great success is, for example, with Kohl's. We also have Neiman Marcus, we have Sachs. But Capital One's focus has been more at the upper end. And we have added, I don't have the count, but we have added several brands that are really good, modestly sized brands on the private label side.

So that business is -- the private label business is pound for pound not as profitable as the pure-branded card business. It's a great business to have in conjunction with a branded card business. But it, for us, has lower returns, but it's a natural part of our business. The top co-brands are something that are very additive to any card program, and we are pursuing those with a lot of interest.

On your second question, the reduction was all a function of re-purchase. But as we talked about earlier, we obviously can't keep it the same level for the next three quarters. We used more than our pro rata authorization this quarter.

Next question, please.

Rick Shane, JPMorgan.

Thanks, guys, for taking my questions. I just want to do a little back-of-the-envelope math about the change to the charge-off guidance for 2017. If we sort of take a rough estimate and say that you have increased the outlook for charge-offs in 2017 by about 30 basis points, and we also make the assumption that the 2015 vintage which is driving this represents somewhere between 15% and 20% of the book. That would suggest in order to get to the change that you have described, somewhere between 150 and 200 basis point change in your loss outlook on that particular vintage.

If you can walk through that math and either agree with it or dispute one of the points there, that would be helpful, so we can understand the magnitude of the change.

Rick, I don't even know how you would get there. Because we've talked about so many factors in the calculation of the allowance growth and mix. And the quarter swaps, there are tons -- they are decomposing that to just the vintage, the composition. I don't know how we can do that without having a 45-minute conversation and getting into a lot more detail than we're going to. So I understand the effort, but I don't see how we could actually get there.

Got it. So, perhaps given the numbers that you have, you could, without -- could you put some context around the change to the 2015 vintage, from where you were at the beginning of the year to where you are now?

I think Rich did put context around it in terms of these are really modest changes.

Rick, I want to go back to, remember, an important part of the change in the guidance for 2017 is about growth in subprime mix. So we're already dealing with only a part of this effect. And then beyond that, this is the composite effect of many different things that are going on.

My point about 2015 was, in many ways, to help, if you pull way up to give a little bit of tangibility to: well, why would we be saying one year's loss outlook is up by a little bit relative to a quarter ago, and the next year's loss outlook is down a little bit relative to a quarter ago? And it's really just because as these things play out, 2015 is coming in a little higher than 2014, and 2016 is coming in a little better than 2015. And that's pretty much the way the math works out. These are all very successful years of origination. I think you are really going to like what we have booked there.

Thanks, Rick. Next question, please.

Bill Carcache, Nomura.

Thank you. I wanted to ask about a different topic. Can you share any thoughts on the current expected credit loss model? And any perspective for the broader industry and for Capital One, in particular, in terms of how you guys are thinking about the impact?

Yes, Bill, thanks for asking about that topic. It's obviously something that we are focused on, and have a lot of people who are starting to build out some of the work. I have a couple of thoughts on that. One, the rules that have come out there allow for a fair amount of judgment and guidance, or judgment. It's more principal-based guidance, as opposed to a set of rules. So I do think that you may see some divergent practices initially, and I think you will see the industry coming together on some topics over time.

So I'm a little cautious about getting ahead of -- putting out estimates and guidance on CCIL, because I do think that we will see people's estimates converge over time, as we all get together on some of the interpretation issues. Pulling up for us, when we look at the effects of CCIL, I think we are not going to be in the business of giving any kinds of estimates on that until we are well into being certain that we know how the standard is going to be interpreted, and what the net impacts are going to be here. So that may not be that you are going to hear from me on this topic until well into next year.

And just to add a little bit of color to that, one of the reasons to hold off is, to the extent that you have literally all of the losses you expect over the life of the loan to be booked up front through your income statement, one could assume that there would be positive benefits to how much capital you need to carry as well. So there's a whole bunch of things that are going to move here in how CCIL gets integrated, and how people think about the combined reserves and capital you need for the business.

Thank you.

Next question, please.

Ken Bruce, Bank of America Merrill Lynch.

Thank you, and good evening. Thanks for squeezing me in. My question is on credit. Sorry, I know that we have been beating this to death, but I'm hoping you might be willing to discuss what you think normalized losses are within the US credit card book? And I don't mean necessarily within your guidance, but just how do you think about normalized losses? And dimensionalize it by prime and subprime, if you could?

Just realizing, we are not there yet, but as you think about what normalized would look like at some point in the future, you have said that auto is back to being normal, and I don't think credit card is. So if we could understand how that looks, that would be very helpful.

Okay. Well, Ken, first of all I want to say and I don't think auto is fully normalized, per se. I think it has inched up a little. I'm talking industry losses now. I think there is a little bit still to go in the normalization of auto losses. And as I said, in many ways, it keeps a little bit outperforming, in a good way, our own expectations of that normalization. But I think there's a little more normalization still to happen.

In the US card business, every card player has a back-book that is at exceptionally low loss level. It's because for years -- a couple of things behind that. And most importantly, anybody in our back-book who survived the Great Recession, not surprisingly, is pretty resilient, and having great credit performance at this point. Additionally, most players didn't grow very much for a whole bunch of years there, so there isn't a lot of front-book effects on people's existing books.

So against that backdrop, our belief is, at some point, there will be some normalization of that credit. And I am reluctant to pick numbers, because that's going to depend on a lot of things. And one of the things it's going to depend on is how much subprime the industry does, for example. If you go back to the late 1990s, there was a, in the middle of an economy that didn't really move at all, credit card losses went up quite a bit just due to a lot of -- and probably subprime is the wrong word. Just a lot of risk expansion and a lot of lending in the card business.

So I've just been around long enough that I find myself, the longer I do this, the more reluctant I am to declare what a normalized loss rate is. But I think there is a net pressure over time that will pull up losses for existing books of all players, including Capital One. We also have one other effect that, at some point, will be beneficial to Capital One, which is the actual -- the other side of growth math, which is the seasoning that, remember, we have talked about the general pattern of credit programs is, they have higher up-front losses.

Now, it depends on what kind of credit program, whether it's an origination, a line increase, what part of the -- you know, subprime is a little different from prime. But if you pull it all together, the general pattern is for front-book programs, they have higher losses early on, and then they gradually settle out over time. So with the big front-book that we have at Capital One, that will at some point become a bit of a good guy in terms of the seasoning affects. I'm not here to predict the exact timing of that, and so on.

So I think over the longer run, when we think about our own book, there is upward pressure on all card businesses and some long-run seasoning that I think will be beneficial for Capital One. The key thing that we focus on, since I don't think that we feel we can absolutely predict the precise through-the-cycle loss rate of our businesses is, is everything that we underwrite, we assume significant worsening. And we focus actually more so than the loss rate itself that we predict, our biggest focus is on resilience.

Because even some of the low-loss programs, this is really the story of banking in general. A lot of time, some of the low-loss programs look great until they don't, and they don't have much of a buffer there. So in all of our underwriting, we are very, in addition to doing all the projections we always do about losses and the whole lifecycle of a vintage, we are very focused on resilience.

So that we have a maximum chance to not only like the business in its full net present value, but to like the business a lot when we get there in the middle of a bad economic cycle. So that's kind of a long answer, but in the context of all of that, we believe that the business we are booking in this front-book is particularly resilient, and frankly, has particularly good economics.

(Inaudible) for a long time, so you can't argue that. Well, thank you. I think it. You've been very generous with your time. I appreciate it.

Thanks, Ken. And thanks, everybody, for joining us on the conference call today. Thank you for your continuing interest in Capital One. Remember, the Investor Relations Team will be here this evening to answer any further questions you may have. Have a great evening, everybody.

Ladies and gentlemen, this concludes today's conference. Thank you for your participation.