發現金融 (DFS) 2018 Q4 法說會逐字稿

Good afternoon.

My name is Celisia, and I will be your conference operator today.

At this time, I would like to welcome everyone to the Fourth Quarter 2018 Discover Financial Services Earnings Conference Call.

(Operator Instructions)

I will now turn the call over to Mr. Craig Streem, Head of Investor Relations.

Please go ahead.

Thank you, Celisia.

Welcome, everyone, to our call this afternoon.

I'll begin on Slide 2 of the presentation, which you can find in the Financials section of our Investor Relations website.

The discussion today contains certain forward-looking statements about the company's future financial performance and business prospects, which are subject to risks and uncertainties and speak only as of today.

Factors that can cause actual results to differ materially from these forward-looking statements are set forth within today's earnings press release, which was provided to the SEC in an 8-K report, and in our 2017 10-K and third quarter 2018 Q, which are on our website and on file with the SEC.

In the fourth quarter 2018 earnings materials, we've provided information that compares and reconciles our non-GAAP financial measures with GAAP financial information, and we explain why these measures are useful to management and investors.

We urge you to review that information in conjunction with today's discussion.

Our call today will include remarks from our Chief Executive Officer, Roger Hochschild, covering fourth quarter and full year highlights and developments; and then Mark Graf, our Chief Financial Officer, will take you through the rest of the presentation.

And after Mark completes his comments, as Celisia said, we will have time for Q&A session.

(Operator Instructions)

So thank you.

Now it's my pleasure to turn the call over to Roger.

Thanks, Craig, and thanks to our listeners for joining today's call.

With this being our year-end call, I want to begin by reviewing full year highlights and key performance indicators on Slide 3, then turn the call over to Mark to review fourth quarter results.

Later in the call, we'll cover our guidance elements for 2019.

We earned $2.7 billion after tax, or $7.79 per share in 2018, generating a return on equity of 25% as we put our capital to work to grow the business and repurchase our shares.

Once again, we generated very strong return on equity, reflecting our unique combination of businesses across consumer lending and payments as well as strong operating performance.

Our performance in 2018 reflected robust receivables and revenue growth, particularly in the card business.

We've continued to invest in brand advertising, in marketing analytics and in superior service.

And together, these investments continue to drive profitable growth.

And of course, we remain focused on providing an exceptional customer experience and are proud to have won the J.D. Power award for credit card customer satisfaction in 2018.

On the subject of how we treat our customers, I want to take a minute to acknowledge the difficult financial situation now faced by furloughed government employees.

To support customers in that situation, we are providing payment holidays to those that ask for flexibility at this time.

Overall, credit performance continues to stabilize.

The impact of normalization is diminishing, and we are experiencing tangible benefits from enhancements to our underwriting and collections strategies.

Our Payment Services segment generated strong volume gains, up 15% in 2018, largely due to the performance of PULSE.

The PULSE team has been successful at winning new relationships and building business with existing issuers by developing creative debit solutions that deliver meaningful value for partners.

Finally, I want to emphasize the importance of investing in our global acceptance footprint and technology.

These are 2 distinct areas, but both are critical to our ongoing success.

In terms of global acceptance, we continue to see a great opportunity to partner with local acquirers to enhance merchant acceptance.

We made significant progress on this in the fourth quarter, signing a number of the agreements in a variety of markets, including the U.K. and Continental Europe.

The universal merchant acceptance remains an important objective as we pursue our longer-term vision of being a leading global payments partner.

And in terms of technology spend, we are continuing to invest in initiatives to drive even better customer experience and competitive advantage.

For example, we're making ongoing investments in machine learning to enable faster and better decisions about how to target our collections strategies and marketing campaigns.

We've begun to see the benefits of these investments in 2018 with positive impacts on loan growth and credit performance.

Now turning to Slide 4. We were pleased to generate strong total loan growth for the year of 7%, with card receivables up 8%.

I alluded to this a moment ago in talking about machine learning, but I am particularly pleased with our growth in new card accounts, even as we continue to tighten credit and brought down the average acquisition cost per account.

Student lending turned in another great year, with organic receivables up 9%, and record originations of $1.8 billion.

Our personal loan portfolio grew 1% in 2018 as we cut back on origination activity by tightening underwriting standards.

Competitive intensity remains high, with new entrants continuing to ramp up originations.

We will maintain our traditional underwriting discipline as we focus on driving growth that meets our return objectives.

Turning to Slide 5. Credit continues to perform in line with our expectations, with seasoning of recent growth and normalization being the key drivers.

We will have more to say about the overall credit environment when we discuss fourth quarter performance in 2019 later into the call.

But as we enter 2019, we feel very good about underlying trends.

I'll now ask Mark to discuss our financial results in more detail.

Thanks, Roger, and good afternoon, everyone.

I'll begin by addressing our summary financial results in Slide 6.

Looking at key elements of the income statement, revenue growth of 7% this quarter was driven by strong loan growth and a slightly higher net interest margin.

Provision for loan losses increased due to the combination of the seasoning of loan growth and ongoing supply-driven normalization in the consumer credit industry.

Operating expenses rose 7% year-over-year as a result of investments in support of growth and new capabilities.

This includes the higher level of incentive payments made in support of the global merchant acceptance initiatives that Roger talked about earlier.

Turning to Slide 7. Total loans increased 7% over the prior year, led by 8% growth in credit card receivables.

Growth in standard merchandise balances drove much of this increase, with a lesser contribution from promotional balances.

We currently expect the promotional balance mix to decline modestly in 2019 and for standard merchandise balances to continue to be the primary engine for growth in card.

Moving to the results from our Payment segment.

On the right-hand side of Slide 7, you can see that proprietary volume rose 6% year-over-year.

In Payment Services, PULSE continues to drive the lion's share of both volume and growth, with volume up 11% compared to the prior year.

Growth was driven by both new issuers as well as incremental volume from existing issuers.

Moving to revenue on Slide 8. Net interest income increased $182 million or 9% from a year ago, driven by higher loan balances and modest NIM expansion.

Total noninterest income grew $11 million, primarily from a 13% increase in loan fee income.

Sales volume grew by 5%, driven principally by growth in active cardmembers.

Adjusted for the number of processing days, sales growth would have been 6%, which is down from 9% on a day adjusted basis last quarter.

Our rewards rate for the fourth quarter was 128 basis points, up 5 basis points year-over-year.

This increase is due to both portfolio mix, which continues to shift toward the Discover it product with a slightly higher average rewards rate, as well as our decision to feature warehouse clubs in the 5% category in the fourth quarter.

As shown on Slide 9, our net interest margin was 10.35% for the quarter, up 7 basis points on both a year-over-year and a sequential basis.

Relative to the fourth quarter of the prior year, the benefit of a higher primary and the expiration of the FDIC surcharge were offset by the impact of an increase in promotional balances, higher deposit costs and higher interest charge-offs.

Relative to the third quarter, the increase in net interest margin reflected a higher prime rate, the expiration of the FDIC surcharge and a mix shift in our liquidity portfolio from cash to investment securities.

Total loan yield increased 45 basis points from a year ago to 12.59%, primarily driven by a 41 basis point increase in card yield.

Prime rate increases and a modest increase in the revolve rate led card yield higher, partially offset by a mix -- a higher mix of promotional balances and higher charge-offs.

On the liability side of the balance sheet, consumer deposits grew 12% as we continued to focus on more stable and cost-effective sources of funding.

Consumer deposit rates rose during the quarter, increasing 12 basis points sequentially and 56 basis points year-over-year.

Deposit betas have increased, though cumulative betas continue to be better than historic norms.

Turning to Slide 10.

Operating expenses rose $74 million from the prior year.

Other expenses were up $40 million, with almost $35 million of the increase due to the investments in global merchant acceptance, which Roger spoke about earlier, and had previewed at the Goldman conference in December.

These were driven by agreements that we signed with new partners as well as existing partners in new territories.

In some cases, acceptance milestones were achieved more quickly than anticipated, and that contributed to the higher level of incentive payments in the fourth quarter.

In marketing, expenses were up as a result of greater brand and digital advertising activity.

Finally, our ongoing investments in infrastructure and analytic capabilities accounted for the increase in information processing costs.

I'll now discuss credit results on Slide 11.

Total net charge-offs rose 23 basis points from the prior year.

The seasoning of loan growth from the past few years and the supply-driven credit normalization continue to be the primary drivers of the year-over-year increase in charge-offs.

Credit card net charge-offs rose 20 basis points year-over-year.

From a sequential perspective, this quarter represented the fifth consecutive quarter of slowing year-over-year increases in card charge-offs.

The credit card 30-plus delinquency rate was up 15 basis points year-over-year, and 11 basis points sequentially.

Our disciplined approach to managing credit with both new and existing accounts has led to continued solid credit performance in the card business.

Private student loan credit performance has also been very strong, with net charge-offs down 17 basis points year-over-year and 10 basis points sequentially.

Personal loan net charge-offs were up 87 basis points from the prior year and 40 basis points sequentially, slightly better than the 50 to 60 basis points we'd expected.

The 30-plus delinquency rate was up 20 basis points year-over-year and 3 basis points sequentially.

Looking at capital on Slide 12.

Our common equity Tier 1 ratio decreased 30 basis points sequentially as loan balances grew.

Our payout ratio for the last 12 months was 93%.

To sum up the quarter on Slide 13, we generated 7% total loan growth and a 25% return on equity.

Our consumer deposit business also posted robust growth of 12%, while deposit rates increased 56 basis points year-over-year.

With respect to credit, our total company charge-off rate, just over 3%, reflects positive underlying trends in card and student loans.

And finally, we're continuing to execute on our capital plan with strong loan growth and capital returns, helping to bring our capital ratio closer to target levels.

Turning to Slide 14.

I want to shift gears and review the key factors that we believe will contribute to another year of strong returns, even as we continue to invest for a longer-term growth.

First, our base case for 2019 adopts the consensus view that macroeconomic conditions remain stable, although our growth plan does contemplate a degree of tightening at the margin.

The economy is growing.

Of particular importance to us as a consumer lender, employment and wages are growing, and consumer leverage remains manageable.

Very simply, we don't see any indication in the data of a turn in the credit cycle.

Second, Discover has a very strong and widely recognized brand, which consumers associate it with value, trust and exceptional customer service, particularly in card.

These attributes remain core to driving profitable growth, and we will continue to invest in brand awareness for our other consumer banking products as well.

Third, we will continue to invest in technology, supporting the deployment of advanced analytics and automation.

This will allow us to make better, faster decisions; drive operational efficiencies in account acquisition, servicing, fraud and collections.

And of course, the thread that we have solved this together is our relentless focus on customer experience.

We recognize that we live in a highly competitive environment, but we've been able to distinguish ourselves by introducing features and benefits that customers value.

Our customers-centric approach has been fundamental to our consistently strong growth in returns.

With that, let's move on to 2019 guidance on Slide 15.

As I said a moment ago, our base assumption for 2019 is a continuation of the current economic environment.

If this proves to be wrong and we see meaningful deterioration in macroeconomic conditions, we would, in all likelihood, pull back on growth and experience an increase in net charge-offs.

Of course, we would also have flexibility in managing rewards costs and operating expenses, which would provide a degree of mitigation.

Looking at the specific guidance elements.

First, we've established a loan growth target range of 6% to 8% based on opportunities that we believe will allow us to continue driving disciplined, profitable loan growth.

Moving to expenses.

We expect operating expenses to be in the range of $4.3 billion to $4.4 billion.

This reflects higher base levels of expenses to support growth in the business as well as continuing investments in the initiatives that Roger and I have mentioned.

With respect to rewards.

We expect the rate to come in between 132 and 134 basis points for 2019, a modest increase largely resulting from the ongoing shift in product mix as we originate all of our new card accounts on the Discover it platform, which has a slightly higher average rewards rate than the predecessor product.

Moving to our outlook for net interest margin.

We expect the full year NIM to come in around 10.3%, with a bias to the upside.

The prime rate increases from last year will contribute positively to the NIM in 2019.

We also anticipate a modest NIM benefit from a lower mix of promotional balances.

Potentially offsetting these benefits would be deposit pricing pressure, wider wholesale funding spreads and modestly higher interest charge-offs.

In terms of the credit costs.

We expect the total net charge-off rate this year to be in a range of 3.2% to 3.4% as a result of the seasoning of a growing portfolio as well as continued, though moderating, supply-driven credit normalization.

So to sum things up, we're pleased with our performance in 2018 and look forward to continued momentum in the year ahead.

One final comment.

Earlier today, we announced internally that Noelle Whitehead would be assuming a leadership role in our student lending business.

I want to take this opportunity to thank her publicly for her outstanding service as a member of Investor Relations team over the last several years.

That concludes our formal remarks, so I'll turn the call back to our operator, Celisia, to open the line for Q&A.

(Operator Instructions) We will take our first question from the line of Mark DeVries with Barclays.

I think you had indicated last quarter that you saw better-than-expected growth in promotional balances in part due to some more effective marketing on your end.

Could you just talk about whether that trend kind of held this past quarter?

And what your outlook is for that into 2019?

And what, if any, impact it may have on your NIM guidance?

Sure, Mark.

Happy to tackle it.

We did continue to see promotional growth play a role in our -- in the growth in our card book over the course of the quarter.

But it was a secondary level of growth, and it was a distant secondary level of growth to standard merchandise purchases.

So I would say we saw a moderation in the component of the growth that was represented by promotional balances.

As we think about looking into 2019, we would continue to see the proportion of growth represented by promotional balances continue to decline modestly.

So in terms of NIM guidance, that's what's kind of baked in there, thought through there.

I think in terms of the 10.3%, with a bias to the upside on the NIM thought just generally, that guidance doesn't include any increases in rates by the Fed, any short-term rate increases.

It's following the forward curve instead of the dot plots.

So that would be -- also be a little bit of upside to margin, if we were to get some of that, probably to the tune of about 4 to 6 basis points for every 25 from the Fed.

Okay.

And anything you can share around what you're assuming around deposit betas in that guidance?

Yes.

We've never talked about our deposit beta, specifically, but I would tell you, cumulatively, the beta's looking -- so the CDs are 33% of the book.

CDs always have a beta close to 1, right?

So we'll put that aside.

But for the indeterminate maturity deposits, the cumulative beta is sitting at 51 as we sit here right now.

So it continues to be well below where you'd expect it to be cycle to date.

I would not expect there'd be a lot of upward pressure on that from market rates.

If there's any upward pressure, it's likely to come from competitors as opposed to a lot of movements in the market.

Your next question comes from the line of Sanjay Sakhrani with KBW.

I guess, on loan growth, I think it's clear your baseline on the economy is that it's okay, and you're seeing opportunities for growth.

But I was wondering if there's any specific attributes in the loan growth that make you comfortable with the resilience through a cycle?

Because I mean, it's pretty robust growth.

Yes, sure, Sanjay.

I would say if I sit and I look at the nature of the growth, the line-weighted FICOs at acquisition in the card side continue to be right at about 730.

So they continue to be very consistent in terms of what we have booked over time.

And as we continue to book these new vintages, we see the card members come on more engaged with the brand, more engaged with the card and driving the types of behaviors that tend to drive long-term profitability.

So we feel real good about that.

And in terms of line utilization, it's not like all these folks are coming onboard and maxing out their lines, so it a good, responsible behavior that we're seeing out of those folks.

The student lending business just continues to be a fabulous product for us.

Risk-adjusted basis, it drives great returns.

We think it's a great introduction of new young people into the fold that could become a part of the broader Discover family as they grow.

The personal loans business, I think we've talked there.

And I would say, we don't expect personal loans to be a giant grower.

As we look into 2019, we do expect it's probably going to be -- balances there will be flattish to maybe a little up, but not a major contributor.

And of course, in student, you're fighting a little bit of attrition from the acquired book, too.

So we would expect card to be the -- to drive the lion's share of the growth in loans over the course of 2019.

Okay.

My follow up is for Roger.

I guess, you got talked about the importance of building out a global merchant acceptance via those initiatives.

Is there any way to sort of assess the payoffs economically as we look at those investments?

Yes.

It's hard to look at that on a merchant-by-merchant basis.

The payoff comes when you reach critical mass of coverage in any given market.

And that benefits not just our own card-issuing business, but also our Network Partners around the world who will see the benefits from that.

So we have a very disciplined investment process that looks at target cost per merchant.

When we can do a partnership with a network in a market, that, of course, lets us build acceptance at a very affordable cost, but that ongoing level of investment is embedded in the expense guidance we've provided for 2019.

Yes, and I think the -- Sanjay, just to pile onto that for a second.

I think the $35 million or so that came in the fourth quarter, I would say largely, we encourage you guys to think about that as a one time.

That's why Roger kind of called it out specifically at Goldman.

Anything else that we see currently is obviously baked into our guidance for 2019.

Your next question comes from the line of Ken Bruce with Bank of America Merrill Lynch.

Let me pick up where you just left off on the comment about the building out of the global network.

Understanding you've got some, I guess, factored into your guidance for next year, is -- do you feel like you're in the early stages of an investment cycle in terms of building out that acceptance?

Is this something that is -- that we're going to be looking at in terms of a longer-term payback?

And maybe just give us a sense as to the kind of where we are in that cycle and you kind of what the payback horizon may look like.

Yes, I would say, it's been a long process of investment, so it's been continuous and it's been something we've been at for many years.

I think the reason it's maybe attracting some attention now is really the lumpiness Mark talked about in the fourth quarter.

But we view building out acceptance both in the U.S. but also in these international markets as critical components to our network strategy, which again, is both monetizing it through our third-party Payment segment, but also driving our proprietary card growth.

Okay.

And maybe a 2-part follow up, just so I can finish on that.

So you would expect the payback to be over a longer horizon, if I can just make that assumption?

You can correct it if not.

And my follow up is just on the -- in the area of competition, we've seen quite a bit of increase in costs related to whether be it acquisitions or rewards, across the board, and I'd like for you to maybe discuss where you think the competitiveness is within the specific segments that Discover is active in and how we should be thinking whether that is, in a sense, getting more competitive or if, in fact, there maybe is some easing which was suggested by the press recently.

Yes.

I mean, we, with our lend-focused model, are always in the prime lending space.

We're not a subprime issuer.

That business is always competitive.

It can go up and down a bit.

But in my 25-plus years in card issuing, there's never an easy time.

You can see from our performance around new accounts that we're booking more new accounts.

Our cost per account is coming down, and all of that is occurring as we've been tightening credit over the last couple years.

So we feel very good about our value proposition and how it can compete.

In terms of overall activity, we believe a lot of the rewards competition has plateaued.

You're not seeing the same pace of introduction of new programs.

So it's going to remain competitive, but we feel good about our product, how it's positioned.

And if you're buying business with big upfront incentives or the highest reward rate in the market, you'll extract a disproportionate share of gamers and transactors, neither fit with our long-term focus on the business.

Your next question comes from the line of Rick Shane with JPMorgan.

Roger, I appreciate your comments related to the shutdown, and just wanted to ask 2 questions related to that.

One is how do you think your customer base indexes sort of more broadly against government employees?

Are you slightly higher concentrated, in line, or lower?

And then second, given that we're now almost 34 days in, which is a pretty significant chunk of time, have you seen any slowdown or change in spending behaviors?

So I'll start with the second one.

We have not seen a change in overall spending behaviors across our portfolio.

Government -- I don't think we over-index for government employees.

We don't have necessarily the employer current for our entire base, but it's not a meaningful percent.

I think it's very important that all of us do everything we can to support the employees in this situation, and so that's why I called it out.

But compared to the programs we've done around major natural disasters, for example, in the past, those have had a much more significant scale.

Your next question comes from the line of Eric Wasserstrom with UBS.

Just 2 questions related to the guidance.

Mark, on the rewards rate, I think historically, the transition from Discover to the Discover it product had about a 2 basis point inflation in the rewards rate.

Is that still true?

Yes.

I would say it's run somewhere over time between 2 and 3 basis points.

I think last year, it was 3 basis points.

So if you think about the -- sort of the guidance range, if you will, what we're kind of calling out is we really don't expect any pressure in rewards other than that continued migration as a greater percentage of the portfolio is made up of it card members.

Okay, great.

And then just on the NCO guidance, I think last quarter, you indicated that on the personal loans component, you expected a change third quarter to fourth, and you came in a little bit better.

But I think the fourth quarter to first quarter of this year, I think that kind of was something like an incremental 40 basis points.

Is that still your outlook?

Yes.

I would say that feels somewhere in the right general zip code.

I'm not going to call it out as a specific number at this point in time, but really, we saw a little bit of goodness in the fourth quarter relative to what we expected.

We installed a number of new technologies in the personal lending business that are helping us really do a better job sniffing out synthetic fraud and really segregating our underwriting a little bit better.

So we are hopeful we can do better than that 5-ish percent kind of guidance we gave last quarter.

But for right now, we're early in the year, so I think I'll stick to something pretty close to that 40 basis points for the first quarter.

Your next question comes from the line of Chris Brendler with Buckingham.

Just wanted to ask on domestic spending and sort of the 6% normalized rate, down from 9% normalized last quarter despite what appeared to be a pretty successful rewards quarter.

Is that just sort of your tougher comps?

Or a little bit of macro slowdown?

Technically, we had a good Christmas, so it was a little bit of a surprise to see that magnitude of deceleration.

Yes.

There are a mix of factors in there.

Part of it is a bit tougher comps.

We booked a lot of new accounts in the fourth quarter last year, and those, of course, come on and activate and start spending quickly.

There was a slowdown in holiday spending towards the back half of the holiday season.

I think you -- that's just not as.

You've seen a number of major retailers call that out.

So those are probably the biggest drivers on the sales front.

Okay.

And a follow up on your NIM guidance.

Mark, does it matter that much what the Fed does?

If the Fed sort of stayed at a more aggressive path it would sort of help out your yields?

If not, potentially deposit costs can start to catch up with yields.

Is that a big swing factor?

Or is that -- doesn't play in fast enough to really matter to the '19 outlook?

No, it's not a giant swing factor.

I mean, if -- a little bit of inside baseball.

If I go back to the fall when we were putting together our operating plan for the year and you looked at the forward curve, it assumed 2 rate increases, 1 in June and 1 in December.

So you would've gotten the benefit of half of a year of one of those increases, and essentially none of a year of the other of those increases.

So if you go to 0, it's not a giant impactor, if you will.

So I'd take you back to that other thought, really sitting back and saying roughly, if we do get them, all in, including the deposit beta assumptions and everything else, you're probably looking at somewhere between 4 to 6 basis points of NIM accretion coming -- flowing through as a result of a 25 basis point move where we'd get one.

Your next question comes from the line of Betsy Graseck with Morgan Stanley.

A follow up to that question.

So 4 to 6 bps up if you get another 25 bps hike.

But if we don't get another hike, then -- I know your guidance is saying 10.3% plus/minus with a bias to the upside.

Is that bias also a 4 to 6 bps range?

No.

It really, Betsy, would depend on too much what happens.

So I mean, I've got -- you know in pulling together thoughts around guidance, we've obviously got 50 scenarios we've run about what if this, what if this and what if this.

I think how you should interpret it is a lot more of those scenarios come out better than 10.3% than come out tougher than 10.3%.

And that's why we're kind of talking about a bias to the upside.

I think at the end of the day, if we see continued mitigation in the rate of increase in delinquency formation, which is kind of pretty flat right now, or if we see charge-off formation continue to moderate, there can be some goodness there.

But in terms of overall, it's going to be a function of market rates, portfolio mix, promo activity, interest charge-offs, deposit betas, funding mix and a whole host of other things.

So it's hard to call.

But I emphasized the bias to the upside for a reason.

Okay.

And then maybe I could just have a follow up on how you're thinking about reinvesting the deposits that you've got.

I mean, in this past quarter, deposit growth was very strong, and I believe the mid-teens or so.

Can you maybe give us some color on how you're thinking about driving that growth rate in a flat rate environment?

Should we expect that, that tails off, slows down?

Or do you think you'll hold that level of deposit growth?

And then how do you think about the reinvestment of that given that loan growth is more in the mid-singles?

Yes.

Clearly, we have multiple channels to use in our funding mix, so it's not as if we'll have to go out and buy assets or accelerate asset growth.

Mark will do his magic on the treasury side.

I think we and others benefited from a big investor move towards cash in the fourth quarter, and so that drove flows up across the industry.

But we do feel good about our ability to continue to grow deposits without posting at the top of the rate tables and having a value proposition very similar to card in terms of differentiated products, service, customer experience, leveraging our brand, and then also, cross-selling into the card base.

So yes, deposits will continue to be the most significant part of our funding mix.

We expect it to grow faster than assets, but that will just, therefore, result in a shift in the overall mix.

Your next question comes from the line of Bill Carcache with Nomura.

The high end of your expense guidance is a bit higher than the low end of your loan growth.

And I know there -- it's difficult to be precise with all of the moving parts, but at a high level, Mark, you mentioned all of the different scenarios.

Maybe could you just speak to, at a high level, your commitment to controlling expenses for the revenue environment and generating positive operating leverage in 2019.

You guys have historically done a good job with deposit operating leverage, but curious if you could speak to that in '19.

Sure, happy to.

I would say the bulk of that investment activity -- I mean, obviously, there is spend ongoing to support the generation of assets in a -- let's call it a BAU-type environment.

But the bulk of that incremental spend is really investment into a lot of these technology platforms that Roger and I spoke about.

Most of these new technologies are domiciled in the cloud as opposed to resident in our 4 walls.

One of the big differences between traditional systems development within your own 4 walls and the cloud is you capitalize the traditional version so it doesn't have a big annual impact on expenses.

Cloud-based development gets expensed.

So a chunk of the increase, really, a big driver of the increase you're seeing year-over-year, is the result of our continued migration to and deployment of a lot of these new technologies that the results we're seeing from them are just really strong, and we feel very good about that.

In terms of just overall, the commitment to expense discipline, no question.

I mean, I appreciate your commentary on the positive operating leverage.

I think it's been a number of years since there was a negative number in front of that despite some really big investment activity on our part.

So I think it underscores the strength of the revenue engine as well.

But we definitively are committed to managing our operating expense base.

We're committed to positive operating leverage, and we unequivocally have leverage in the model, as I called out in my prepared remarks, that if we see something in the environment start to change, we've got leverage in marketing dollars, we've got leverage in rewards costs, we've got leverage in a lot of different line items in the P&L that we won't hesitate to pull those levers if it's the right thing to do.

That's pretty helpful.

If I can follow up quickly on credit.

On an annual basis, we saw your provision growth exceed your loan growth by incrementally larger amounts in '15, '16 and '17.

But the excess of provision growth over loan growth actually shrank in '18, and this trajectory is consistent with some of the favorable credit trends that you highlighted in your prepared remarks.

As we look ahead to 2019, is it reasonable to think or to expect that, that excess of provision growth over loan growth should continue to compress modestly?

Yes, I'm going to kind of stay away from answering that one specifically because I don't want to give provision guidance.

What I would say is, we have called out that we continue to see the rate of increase in charge-offs in the card business moderate for 5 consecutive quarters.

We obviously set our reserves and provide on a quarterly basis based on a whole number of factors that we see at that point in time.

But the fundamental underlying, underpinning piece that is the biggest component of it, obviously, is what are the trends we're seeing in credit.

And those trends, with normalization slowing, continue to feel pretty good.

But in terms of actually calling on a forward basis what provision's going to be, I'm going to stay away from that one.

Your next question comes from the line of Bob Napoli with William Blair.

New card growth, 11%, and cost per account, down 6%.

A pretty impressive combination.

Is the new card growth in acceleration?

And what is going on with net card growth?

Is the runoff picking up?

What is driving those really, really good metrics in card growth and cost per account?

And then are those, I guess, the same quality of accounts?

Are you seeing any faster runoff of accounts?

Yes.

So in terms of -- I would separate the 2. First, we have not seen any change in attrition.

And given how we deliver on customer experience, our attrition tends to be the lowest in the industry.

In terms of what's driving our card acquisition performance, it really is a lot of those investments around technology, in particular, next-generation analytics and modeling as well as just how well the value proposition competes with millennials and younger consumers, especially.

So we're very pleased to see that.

And again, I'll repeat, all of that is in the context of continuing to tighten credit around our new account underwriting.

Great.

A follow-up question.

Just in line with the 2019 growth, Mark, one area you didn't give any color around, and not that you have in the past, is the noninterest income revenue growth.

And that's kind of been a low single-digit revenue growth item.

Is that the continued expectation, is kind of low single-digit growth of noninterest revenue?

Yes.

I don't think we've historically guided on noninterest revenue, Bob.

And so I'm going to kind of stay away from that one.

It's not really a giant driver of our -- component of our P&L.

Your next question comes from the line of Chris Donat with Sandler O'Neill.

Mark, had sort of a clarification as you used the expression supply-driven credit normalization a few times on the call.

Just want to see if you could expand on that one.

And then if you want to get philosophical, maybe also give us your view on what -- if, at some point, we ever have an end in to the credit cycle, what might cause that?

Well, the first part, that's easy to do.

I would say the supply-driven credit normalization, if you think about the components that would drive an increase in charge -- delinquencies and charge-offs, you really have an incidence component.

In other words, what percentage of the portfolio comes under stress.

And then you have the severity component, which is once an account does come under stress, what's happened to the balances?

Are they bigger now than they were a couple of years ago?

And I would say the driver cycle to date in this has really been -- in our case, has really been we have seen severities increase much more so than we've seen incidence rates increase.

So if you think about it going through the crisis, consumers delevered, either by choice or otherwise.

And so for a number of years coming out of the crisis, they carried really low levels of leverage.

A few years back, consumer credit availability started creeping back in.

Consumers started levering up.

So while we're not seeing a greater percent -- a much greater percentage of the book come under stress, when somebody does come under stress, they're carrying more debt.

Therefore, the severity of the charge-off is bigger.

We're referring to that as supply-driven normalization because supply of consumer credit, that's been really been the primary driver of that.

Speculating as far as what's going to be the driver of the next recession, there is a lot smarter people than me that you could ask who could give you a better sense on that one.

I guess, what I would say is looking at the data, seeing everything we see, we don't see it as a consumer-led issue.

It feels more geopolitical, global macro, something like that as opposed to an over-leveraged consumer or a consumer asset bubble.

Got it.

I appreciate that one.

And just as a follow up, I'll ask a very pedestrian one of, the other expense line on the income statement.

It's just a little higher than normal.

Is that where we saw some of the merchant acquisition?

Or your cloud-based investments show up in that line?

Or are those elsewhere?

It's the global network stuff that we spoke about earlier on the merchant side that is showing up there.

Your next question comes from the line of John Hecht with Jefferies.

Well, all of my questions have been asked, so I'm going throw out a painful topic, Mark, and it's related to CECL.

Sure.

I think a specific question is, I believe you're all supposed to start running parallel models now.

And I guess the specific question is, do you have enough, I guess, insight to all the elements of CECL to begin running that?

If not, what are the uncertainties around what the modeling would entail?

And if so, do you have any comments on that at this point?

I knew the question was going to come up, so no worries about the painful ask.

Not a problem.

I guess I would say one of the places a lot of that technology that Roger and I have alluded to a few times ago has been put to work in is in the CECL modeling, trying to break away and come up with some machine-learning approaches, really starting to think about every way we can to start slicing and dicing our portfolio.

So we are in the process of going through what I would describe as a giant Monte Carlo simulation.

The standard itself is very broad in terms of how you can think about it and how you can define certain things.

So we're doing a Monte Carlo simulation to really kind of figure out what the right answer for Discover is, right?

What's going to produce essentially the best, least volatile answer that most accurately reflects the loss content in the portfolio.

So I would say the technology is largely in place.

Now we're in the process of running these Monte Carlo simulations to figure out how exactly we want to think about this going forward.

So parallel, yes.

At some point in time this year, we will go to a full parallel.

There's no question about that.

And as we said at that point in time, once we have a sense of what that CECL reserve's going to look like, we feel like we have a disclosure obligation, we'd put it in front of you.

I would say right now, I would just underscore what we've said before, and that is in a CECL environment, reserves will be higher and more volatile for consumer loans.

That's just the nature of the beast in terms of the way the standard is set and it would work.

On the positive note, we continue to interact heavily with the FASB, along with a number of our peer institutions who are also actively engaged.

There is a proposal out there to run a lot of the volatility created in CECL through AOCI as opposed to through the income statement.

And we're hopeful that the powers that be will see the wisdom in that instead of creating a truckload of EPS volatility that doesn't reflect the underlying trends in the business.

Your next question comes from the line of Moshe Orenbuch with Credit Suisse.

And most of my questions, actually, also have been asked and answered.

And I think that it is kind of interesting that you've been talking for a couple of quarters about improvement in account acquisition costs, and I think it's particularly notable that you're able to do that with -- without increased reliance on promotional balances.

And so maybe it's not a fair question, but it seems like the entire industry is trying to do that.

What -- is there anything you can tell us that you're able to -- any reason that you're able to do that and others aren't?

Yes.

One thing we've always highlighted is the differentiation we get from our proprietary network.

And so I do think you see 2 issuers that have proprietary network really stand out in terms of growth, and both of us tend to be conservative on the credit side.

The other thing I would say is, it's just that relentless focus on the customer.

We've won the J.D. Power Customer Experience award 4 of the last 5 years.

It's not just the great 100% U.S.-based customer service.

It's the mobile app, it's the rewards, it's the new features and benefits.

So we look carefully at our targeted customers and are always thinking of new ways to add value.

Got it.

And maybe just from a credit standpoint.

Since you've kind of given some expectation that losses will be higher -- the rate of increase on the personal loan side, while it's a smaller piece of the portfolio, is going to be higher than your overall expected rate of increase.

I guess -- I mean, the arithmetic conclusion is that you're expecting the increase in losses on credit cards to be less than that.

I mean -- and so -- I mean, I think that -- any kind of thoughts or additional insight you can give us there?

No, I would say mathematically, I agree with your conclusion, Moshe.

It really -- we've seen some pretty big increases in the personal loan book, and I think we've kind of called that out and talked about that one, as you know, pretty extensively in the past.

We're hopeful that we can actually produce a better result than what we called out over the course of the last couple of quarters in terms of looking into '19 there, but we're early in the year.

So I'm not going to declare victory.

But yes, the rate of increase in the charge-offs in the card book continues to moderate.

We feel very good about that.

Student loan book, you can see in the supplement, the statistics there is performing exceptionally well.

So -- and unless something else pops up that changes the trends, it really does feel that both the card and the student businesses are performing really well from a credit perspective right now.

And your next question comes from the line of Don Fandetti with Wells Fargo.

So Roger, are there any competitive implications from the Fiserv First Data acquisition in terms of PULSE on PIN debit (sic) [Internet PIN Debit]?

And then secondarily, how do you think about that business?

Is that a strategic asset?

I know it sort of leverages into rewards and debit cards and plays into checking.

But have you ever thought seriously about trying to monetize that?

Yes.

So first, we don't discuss M&A, so acquisitions and divestitures.

But in terms of the merger, I'd say it's probably too early to tell.

Both companies are good partners of us.

We compete in some parts.

We partner in others.

And so I think we'll watch that very carefully.

In terms of the PULSE business, coming off a very strong year of performance.

And clearly, as we look at our Payment segment being able to offer both credit and debit capabilities out there is very important as a network.

So we feel very good about the payments assets we've assembled, both to the PULSE acquisition as well as Diners, and are working hard to continue to increase the share of profits that come from payments.

Your next question comes from the line of Vincent Caintic with Stephens.

Just a few quick follow-ups.

So just first on -- a follow up on the global merchant acceptance.

So it's great to see the partnerships that you've put together.

Just kind of wondering, when we think about the medium term, what are your thoughts about increasing the merchant acceptance in terms of what activities you plan to do there?

Should we expect more partnerships, different products, other investments?

Yes.

I would say it's a continuation of a multiyear journey.

So again, getting a lot of attention because of lumpiness in a single quarter, but we have been working -- if you go back and look at the stream of announcements, we've been working with merchant acquirers and markets outside the U.S. for many years.

The one part I would say we're particularly focused on are our partnerships with networks around the world, which brings us both broad acceptance in different markets but also volume as they leverage our account number ranges, our chip specs.

And so it helps strengthen the overall network.

And so that's a strategic thrust that you can expect to continue.

Okay, great.

And then just one more, shifting to the private student loan market.

So noticed that the credit trends have gotten better year-over-year in 2018.

They got progressively better over the course of each quarter in 2018.

Just wondering if there's anything that's -- in particular that's driving that, and what should we expect for 2019?

I would say, you've had a growing portfolio at a pretty rapid rate over the course of a couple of years.

And as that portfolio seasoned, it was getting to a maturation level.

You were seeing a bit of uptick in the credit statistics and the charge-off rates and the like.

As that portfolio begins to -- as the growth there begins to moderate a little bit, I think you're seeing some of the impact of that.

I think the other key piece of the puzzle is really just the disciplined nature of the underwriting that's continued to go on there.

You've got an average FICO in that book in the order of 750 right now, so it's a full 20 points higher coming on the books than the average card account is.

And just to remind all of you who don't live in the FICO world normally, 20 FICO -- every 20 FICO points doubles default risk.

So a 750 defaults at half the risk of a 730 as a starting point on addition to that.

So I would say, that's a key piece of the puzzle as well.

And obviously, not to try taking credit for brilliance, you also have a really strong economy out there.

These kids are coming out of school and they're able to find jobs, right?

So that's a piece of the puzzle as well.

So I'd say all of those are contributing factors.

Your next question comes from the line of Dominick Gabriele with Oppenheimer.

The new loan growth guidance is a bit lower than what we saw in '19.

Can you talk about how much of this is related to gas prices and non-gas price impacts?

Because the impact be about maybe 2%.

And then are there any other categories that are standing out within spend that you see potentially slowing down?

Or is this just really related to you tightening those new card acquisitions?

You bet.

So I think there's a couple of pieces embedded in there.

We do over-index as a card to gas.

It tends to be on the order, these days, of about 5% of our sales.

That's down from about 10% of our overall sales when gas prices were higher.

So there definitely is a muting impact associated with the decline in gas prices.

There's no question about that.

Certainly, an element of it reflects just the tightening around the margin that Roger referenced earlier in terms of our credit standards in terms of booking new accounts.

And then the third piece of it, obviously is, look, personal loans, as we've talked about, we expect to be pretty much a flat to a very modestly growing book.

So -- and that's the book that over the last couple of years has been a significantly higher grower.

So I think taking that growth component out of the mix as well also has an impact.

So put all those together and that's kind of how we're triangulating on that range.

And just really quick.

Can you talk about if you expect any impacts in the next few years in your student loan originations and loan growth, given that Navient is kind of coming up to that here with a new in-school student lending product.

And then maybe pushing a little harder on the refinanced loans, could you talk about the industry dynamics there?

Yes.

I mean, I think we tend not to focus on any specific one competitor.

So I would say in general, the refi volume -- not them, but just the whole growth of student loan refi, has had an impact on our payment rate.

And so that's something we're watching closely.

We've still been able to grow through that, but it's had a noticeable impact.

And so I think we'll see how sustainable that is, both a, as those companies, at some point, need to focus on profitability; and then, b also, as rates have risen.

So refi does have an impact, but we are continuing to grow through it.

Your final question comes from the line of Jill Shea with Citi.

So maybe just quickly on credit quality.

In the past, you've mentioned a portion of the loan loss reserve bell that's related to the seasoning of accounts versus the normalization of the back book.

I was just wondering if you could give us an update there and what you're seeing in terms of trends?

Sure, I'd be happy to.

In the fourth quarter, this is in a sniper rifle kind of estimate, it's more of a ballpark estimate.

But I would say about 2/3 of the provisioning was related to new accounts.

About 1/3 of the provisioning, ballpark, was related to the seasoning of the back book, again, reflecting that moderating pace of normalization taking place across the industry that we referenced earlier.

And at this time, there are no further questions.

I like to turn the call back over to Craig Streem for any additional or closing remarks.

Sure.

Thanks, Celisia, and thanks, everybody, for your questions.

And we're available for any follow-ups that you might have.

Thanks.

Have a good evening.

This concludes today's call.

You may now disconnect.