發現金融 (DFS) 2021 Q4 法說會逐字稿

Good morning. My name is Britney, and I will be your conference operator today. At this time, I would like to welcome everyone to the Fourth Quarter and Full Year 2021 Discover Financial Services Earnings Conference Call. (Operator Instructions).

I would now like to turn the call over to Mr. Eric Wasserstrom, Head of Investor Relations. Please go ahead.

Thank you, Britney, and good morning, everyone. Welcome to this morning's call. I'll begin on Slide 2 of our earnings -- as defined in the financial section of our Investor Relations website, investorrelations.discover.com.

Our discussion today contains certain forward-looking statements that are subject to risks and uncertainties that may cause actual results to differ materially. Please refer to our notices regarding forward-looking statements that appear in our fourth quarter earnings press release and presentation.

Our call today will include remarks from our CEO, Roger Hochschild; and John Greene, our Chief Financial Officer. After we conclude our formal comments, there will be time for a question-and-answer session. (Operator Instructions)

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

Thanks, Eric, and thanks to our listeners for joining today's call. I'll begin by reviewing the highlights and key metrics for the year, then John will take you through the details of our fourth quarter results and our perspectives on 2022.

2021 was another year of unique challenges related to the pandemic, and I'm very pleased that, once again, the Discover team was able to successfully execute against our business priorities in a fluid operating environment. This was evident in our fourth quarter results, which were the capstone to an outstanding year. For the fourth quarter, we earned $1.1 billion after tax or $3.64 per share; and for the full year, $5.4 billion after tax or $17.83 per share. These results underscore the strength of our differentiated model and were achieved as we continued to make meaningful enhancements to our capabilities and invest for future growth.

Let me share a few examples from this past year. Throughout 2021, we continued to make advancements to our data and analytics platform, enhancing our capabilities in areas including targeting, collections and fraud detection. We also made investments in machine learning to provide faster and better insights to improve customer personalization. And we continue to modernize our infrastructure and build out our hybrid cloud platform.

We also opened a new customer care center in Chatham, a vibrant African-American community on Chicago's South side. Once fully operational, the center will provide nearly 1,000 full-time jobs. Our Chatham Center challenges the traditional notions of corporate site selection, has helped us connect with a talented pool of diverse candidates and suppliers, is transforming how we approach diversity, equity and inclusion. And it is already performing at an industry-leading level. We hope our commitment to Chatham will serve as a springboard for further economic development in other areas that have long been denied opportunity.

Slide 4 of our presentation captures another important element of our results, which was our pivot into new account acquisition as the economic recovery took hold in late 2020 and early 2021. In the face of intensifying competition, our value proposition of cashback rewards, no annual fee and industry-leading service remained very attractive to consumers. The strong level of card acquisition contributed to our return to loan growth over the second half of last year.

In payments, we continue to expand our business, increase network volume and establish new strategic partnerships. We expanded global acceptance and announced new network alliances in Portugal, Bahrain, Jordan and Malaysia that will benefit us as cross-border travel recovers. We remain committed to building out our international acceptance, and we'll continue to make investments to expand our reach.

Our record earnings through the year generated significant capital, which we continue to put to good use. In addition to our investments in acquisition, brands, technology and people, we also returned significant capital to shareholders through dividends and buybacks by repurchasing $2.3 billion of common stock and increasing our quarterly dividend by 14%.

As we look forward into 2022, I'm very optimistic about the trajectory of our business. While macroeconomic conditions created strong tailwinds this past year, we acted on opportunities to strengthen our business, actions that will drive long-term value this year and beyond. We start the new year in an excellent position, and I'm confident that our integrated digital banking and payments model will continue to create long-term value for our shareholders and customers.

I'll now ask John to discuss key aspects of our quarterly financial results in more detail.

Thank you, Roger, and good morning, everyone. As we review our fourth quarter results, I echo Roger's point, that the actions we took last year positioned us for strong performance in 2022 and beyond. I'll begin with our financial summary on Slide 5.

Our strong fourth quarter results were characterized by accelerating receivable growth, provision leverage and increased investments in marketing and brand. Revenue net of interest expense increased 4% from the prior year. Excluding a $139 million unrealized loss on our equity investments, total revenue was up 9%. Net interest income increased 4%, driven by growth in average receivables and an 18 basis point improvement in our net interest margin, which was sequentially flat at 10.81%. Our NIM trend reflects the continued benefit from decreased funding cost and lower interest charge-offs, though these were partially offset in the fourth quarter by a higher mix of promotional rate balances.

The growth in receivables was largely driven by card, which was up 4% year-over-year and 6% sequentially. The primary drivers of year-over-year growth were continued strong sales volume and significant new account growth throughout 2021, which was up 23% year-over-year and 13% versus 2019.

As has been the case for most of last year, a significant portion of the benefits from strong sales and new accounts was offset by the sustained high payment rate. The payment rate leveled off during the quarter but remains approximately 500 basis points above pre-pandemic levels. We currently expect that the payment rate will decline slightly over the course of 2022. However, our expectations for card receivable growth is robust, as I'll detail in a few moments.

Our student loan portfolio also contributed to our growth. Organic student loans were up 4% over the prior year, benefiting from the return to in-person learning in 2021. We continue to gain share in this product and are well positioned for continued organic expansion.

Personal loans decreased 3% year-over-year, mainly driven by high payment rates, but were sequentially flat as we returned our underwriting criteria to pre-pandemic levels.

The second significant driver of our revenue growth was higher net discount interchange revenue, which increased $103 million or 43%, driven by a 25% increase in sales volume year-over-year. The strong volume has continued into this year. Sales are up 24% through the first half of January.

One significant item that relates to our net discount and interchange revenue is our rewards cost. And having covered most of our key revenue drivers, I'll point your attention to the reward rate reflected on Slide 8.

We continue to benefit from strong card member engagement with our cashback rewards program. our rewards costs increased versus last year on higher sales volume. However, the reward rate declined 3 basis points year-over-year and 9 basis points sequentially. This reflects the benefit of our integrated model and our discipline in managing the program while delivering substantial value to card members and merchant partners.

For the full year 2021, our rewards rate was 1.37%, up 2 basis points from the prior year. Consistent with the historical trend, we expect about 2 to 4 basis points of annual rewards cost inflation, driven mostly by shifts in mix.

Now I'd like to spend a moment speaking about expense trends on Slide 9. Total operating expenses increased $34 million or 3% year-over-year. Focusing on the most significant items here.

Marketing expense was up $112 million as we continue to invest in new account growth and brand marketing with the launch of our new media campaign, which went live across all channels in the quarter. This pushed our marketing expense towards the top end of our previously guided range.

Employee compensation was down $5 million year-over-year, driven mainly by a $26 million charge last year. Excluding this, our comp expense was up 4% year-over-year, driven largely by a higher bonus accrual. Information processing was down $73 million year-over-year and professional fees increased as a result of higher recovery fees.

Moving to credit on Slide 10. The net charge-off rate improved to 1.37% in the quarter, a decrease of 101 basis points year-over-year and a 9 basis point improvement from the prior quarter. Net charge-off dollars were down $218 million from the prior year and decreased $12 million sequentially.

Strong credit performance continued across all products. Card net charge-offs were down 113 basis points from the prior year and personal loans were 158 basis points lower. Student loan charge-offs increased slightly but remained very low at 0.8%.

Moving to the allowance for credit losses on Slide 11. Our reserve rate continued to decline, dropping 38 basis points to 7.3%. Two factors contributed to the decrease in reserve rate. First, we released $50 million from reserves during the quarter, driven by the continued strong credit performance of our portfolio and the relative stability of the macroeconomic outlook. These factors were partially offset by the 5% increase in total loans from the prior quarter. Our future reserves will be dictated by our portfolio credit trends, our receivable growth and any changes to our macroeconomic assumptions.

Looking at Slide 12. We remain extremely well capitalized and above our 10.5% target with a common equity Tier 1 ratio of 14.8%. We continue to demonstrate our commitment of returning capital to shareholders as we executed on our share repurchase plan and bought back $773 million of common stock in the quarter and paid a dividend of $0.50 per share.

Looking at funding, average consumer deposits decreased 3% year-over-year and declined 1% sequentially. This sequential decline was driven by a 5% decrease in consumer CDs, while savings and money market deposits increased slightly. We managed our excess liquidity down throughout 2021 and finished the year with consumer deposits representing 68% of total funding. We will continue to target 70% to 80% of funding from [this source].

Moving to Slide 13, where we'll provide some perspectives on 2022. We entered the year in a very strong position, and our outlook reflects this. We expect loan growth in the high single digits. This view is based on current expectations of sales trends and the contribution from recently acquired accounts, combined with a very modest decline in the payment rate. We believe this view of payment rates substantially derisks our loan growth forecast.

We expect our NIM rate to be relatively in line with the full year of 2021 and with quarter-to-quarter variability. We expect to benefit from higher loan yields with rising interest rates. This may be offset by other factors, including a higher mix of promotional rate balances, some degree of credit normalization and higher deposit rates, which will be subject to funding needs and competitive dynamics.

Turning to expenses. We expect our total GAAP expenses will increase at a mid-single-digit rate this year. We'll continue to invest for growth as we see profitable opportunities and currently expect that our marketing investments will be above 2019 levels. Outside of marketing, we expect operating costs to increase at a low single-digit percent level, reflecting disciplined expense control. Our commitment to positive operating leverage over the medium term remains a priority.

We expect net credit losses will average in the range of 2.2% to 2.6% for the full year. As credit normalizes from historically low levels in 2021, we expect net charge-offs to increase sequentially over the course of the year.

Lastly, we remain committed to returning substantial capital to shareholders through dividends and share buybacks. As of this week, we had approximately $780 million remaining on our share repurchase authorization that expires at the end of March and expect to announce a new share repurchase authorization next quarter.

In summary, we had an excellent fourth quarter and full year with accelerating loan growth driven by robust account acquisition and strong sales volumes, excellent credit performance and a reduction in the reserve rate, disciplined management of operating cost and sustained return of excess capital to shareholders.

I'm exceptionally pleased with Discover's execution against our business priorities in 2021. Our value proposition continues to resonate with consumers. We prudently invested for growth, resulting in significant new accounts growth and strong sales. We continue to optimize our funding mix and actively manage core deposit costs. These actions, and the improved macroeconomic outlook, have positioned us well.

With that, I'll turn the call back to our operator, Britney, to open the line for Q&A.

(Operator Instructions) We'll take our first question from Moshe Orenbuch with Credit Suisse.

Congratulations on pretty strong numbers here. Maybe just focus on the account growth, it was up 11%. Could you talk about like how that compares in terms of just raw numbers of accounts to where you would have been pre-pandemic? And the cost? And anything you've done kind of from a difference in channel or credit box to get there.

Yes, we don't disclose the number of new accounts. But what I would say in terms of credit, our card credit policies are pretty much on top of where they were pre-pandemic. Costs have gone up since the low levels that we saw during the pandemic as we sort of kept marketing and a lot of others pulled back. But I would say, as we look forward into 2022, our expectation is that the cost per account will be roughly where it was pre-pandemic.

And just as a follow-up, very pleased to hear that the commitment to operating leverage and the numbers you put out there would suggest kind of an upper single-digit revenue growth rate and a mid-single-digit expense growth rate. Could you talk about, even if it's unlikely, what would happen to your expense expectations if revenue growth just didn't materialize? Like how would you make adjustments there?

Yes. Hey, Moshe, I'll take that one. So we'll react based on the company performance and opportunities that we see in the marketplace. So over the past couple of years, I feel like the team has done a really, really good job in terms of making calls, in terms of expense allocations. We'll continue to do that to make sure that we're positioned for growth and marching towards the positive operating leverage that we talked about.

We will take our next question from Mark DeVries with Barclays.

Could you give us a little more color about what you're seeing in credit right now? I mean, I think the guidance around charge-offs calls for some pretty meaningful normalization in 2022. And what kind of gets you to the low end of that guidance range and what gets you to the high end?

Great. Mark, thanks for the questions. So what we're seeing in the portfolio is really, really strong performance. I think the numbers in the quarter demonstrate that. We are seeing some difference between higher FICO and lower FICO account performance. Overall, all within expectations. So what we're projecting here is a slow normalization of credit. And in terms of the guidance that we reflected here, we've got pretty good line of sight to the first 6 months, just as the accounts will go through their normal roll rates and gives us an ability to predict quite accurately.

Once we get out beyond 6 months, we rely more heavily on our models, and the models themselves are built with a number of assumptions. '21 was, frankly, was a tough year to call based on how we had designed our models and what actually happened in the portfolio. So what we decided to do is give, from my standpoint, a relatively broad range that we think, over time, we'll be able to tighten.

And in terms of lower end versus higher end, strong portfolio performance. The slight difference that we saw between the higher FICO and the lower FICOs, that continuing to kind of roll out as we expect. And a positive macro environment should bring us towards the lower end.

Just a follow-up on that. On the model that you use, is there some element of just mean reversion when charge-offs are this low relative to kind of normal that will push your estimates higher, even if kind of all the macro drivers seem like they're more of a tailwind than a headwind?

Yes. There is a level of mean [revision] in there in the second part of the year.

And just one other piece of, I'll say, information. So you can think about the breakout, at least this is how we thought about it, of charge-offs, the predominant piece in the second half of the year. So you can think 45/55 split between first half and second half. And we'll continue to update that over the year.

We will take our next question from John Hecht with Jefferies.

Just thinking about NIM, understanding you guys are guiding for a relatively flat NIM this year, but even thinking this year and beyond, given that we're in a rate hike cycle. You're in -- I guess the last cycle was 2015 to 2019, and your card yields went up about 50% of the Fed funds changes and prime changes, and your cost of capital -- or your deposits went up a little less than that.

I'm just wondering as we go into this rate cycle, should we expect any differences in that kind of -- in that, call it, NIM input range? Is there any mix shift in the deposits that will cause the betas to be different there? Or is there any different policies with respect to 0-balance transfers that we should think about? Just in terms of the quarter-to-quarter fluctuations as the government or the Fed raises rates.

Yes, it's actually a pretty complex question there. So rather than try to hit each of the elements, I'll give you a view in terms of how we thought about NIM for 2021.

So we do expect some impact to NIM from credit normalization. We also are expecting a higher mix of BT and promotional balances, which will also impact net interest margin. We do get nice fees from that, but certainly, net interest margin will be impacted. The revolve rate, as the payment rate starts to normalize, we'll see some benefits there.

Fed funds rate changes, we plan for 2, 2 in the year. If there's more, that will create some upside. And in terms of impact there, you could expect, per Fed change, somewhere between 3 and 5 bps on a total year basis on NIM, depending on timing.

And then funding, we're still going to see some funding benefit from the actions we did to optimize our debt stack in '20 and '21. So from that standpoint, it gets us to about flat, but there's a lot of moving pieces.

In terms of deposit pricing, there's still a wide gap between where we're priced and the brick-and-mortar banks are priced. So somewhere close to 45 basis points. Now the digital banks, most have increased deposit pricing, about 10 basis points over the past 3 to 4 weeks. We're a lagger on that. At some point, our funding needs and competitive dynamics will be such that we'll take a look and make appropriate changes, and we'll do that throughout the year.

So the deposit betas, specifics around that, I would rather think about competitive dynamics and funding needs in order to get a view on how we're going to price deposits on the up cycle.

A follow-up question, I guess unrelated, is, you're in the zone of where you were on your allowance levels as we entered post-CECL implementation and pandemic era. Are we at a point now where you'd kind of say this is the new base for ALL, ex changes in our economic forecast? Or is there any kind of further room for allowance bleeds tied to increases during the last couple of years?

Yes. So we go through a pretty robust process every single quarter to make sure that our reserves are fairly stated under GAAP. Now when we did CECL day 1, we had -- we were seeing charge-offs in the low 3s, and we ended up posting a reserve rate of, I think it was 6.09%. So as we look at where we are today, where it's at around 7.2%, assuming strong credit performance and a positive macroeconomic outlook, my sense is that there is some opportunity to take that reserve rate closer to day 1.

We will take our next question from Rick Shane with JPMorgan.

I'm actually curious if there's some sort of 80/20 rule that impacts payment rate. What I'm curious is if you guys have done any analysis on revolve versus transaction behavior differing by spending categories specifically, like everyday spend versus large episodic spend. And I'm wondering if what we're seeing in terms of payment rates with all the other factors is being impacted by BNPL.

Yes. I wouldn't say there's an 80/20 rule on payment rate. So quite honestly, as we looked at it, in 2021, we didn't call the curve right. And so what we did in the '22 guidance is assume a very modest decrease in payment rate. We felt like that derisked our loan growth.

In terms of kind of behaviors between revolvers and transactors, that payment rate's been relatively consistent. The one difference in our portfolio is the new accounts that we originated in 2020 tended to be higher FICOs. So more of those tend to transact versus revolve. But we'll see as the liquidity ends up getting used in the overall economy and what that does to payment rates.

So we've modeled it multiple different ways. There's really no standard rule of thumb. It's frankly somewhat dynamic in terms of what we're seeing month-to-month and quarter-to-quarter.

And just to build on that. We look pretty carefully merchant, by customer segment. There's nothing to support a view that buy now, pay later is having any impact on payment rate.

And we will take our next question from Sanjay Sakhrani with KBW.

First question, I guess, for Roger, just on the competitive backdrop. Obviously, we've gotten bank earnings, and a lot of the large banks are talking about investing more in marketing, and you guys have some pretty aggressive growth targets. How should we think about sort of the risks to that target given the competitive backdrop?

Good question, Sanjay. As I mentioned earlier, our forecast for next year is cost per account relatively flat to where it was before the pandemic. And so I think this sort of focus on competition is a little bit overblown. The card business is just always competitive. You have big players with good capabilities. Each issuer out there has their set of products, set of channels and I think some natural limits on how much money they can be put to work effectively.

So I don't see it as a particularly high risk to our 2022 growth forecast. And we have a very differentiated product, we're seeing good benefits; on the acquisition side, from our investments in analytics. So we feel good even in the current environment.

And I guess I have one for John. The noninterest income revenue line has done quite well over the course of this year, even excluding -- obviously, excluding the investment gains. Could you just talk about what kind of growth we should expect in that line going forward?

Yes. So the big driver there was obviously discount and interchange revenue, which was up 28% year-over-year and then after rewards cost 43% in the quarter. So super strong growth there, which that will generally run consistent with sales. And we're expecting sales not to stay in the mid-20s, but kind of come down kind of strong double digits, and then into weaker double-digits by the fourth quarter. I don't know if that's right or not.

It seems like there's been a lot of benefit from the new account acquisition and also where our card has been positioned in people's wallets. So that's been positive. There's also the factor that people are using less cash and charging more. So that will continue to benefit sales, interchange and -- net discount and interchange.

In terms of the other items, the amount of cash in the economy actually helped cash advance fees, which was positive. And our expectation is that we'll have, outside of discount and interchange growth, growth fairly similar to net interest income.

And we'll take our next question from Betsy Graseck with Morgan Stanley.

Can you hear -- Hi. Can you hear me? Okay.

One, maybe it's a little bit of a theoretical question, but I'm just trying to understand how you were thinking about borrower capacity, the borrowing capacity of your customer set relative to pre-COVID. And I'm asking a question because jobs more available, wages rising, so that seems like they might have more capacity to borrow, but then we've got inflation increasing. So yes, how do you think through those things?

It will, of course, vary by segment. But certainly, households that are seeing rising incomes will have a greater ability for debt service. And for many households, that's how they determine how much debt they take on. It's complicated a bit by some households still having pent-up savings from strong earnings and not many opportunities to spend. And of course, other costs going up, whether it's child care or day-to-day expenses, on one hand, will drive increases in sales. But on the other hand, decreases disposable income for debt service. So a mix of factors, but we, net-net, would expect strong consumer demand for credit next year.

Okay. And then as I'm thinking about the other legs of the loan growth platform here, student loan and personal, could you talk through how you see those drivers impacting the 2022 growth guide that you've got?

Sure. So student loans, we feel really good about where we're positioned in the market, our products, our brand, our ability to take share. What's a bit the wildcard is just enrollment. And I think enrollments were down year-over-year, which was a surprise to, I think, the entire higher education industry. There's a bit of a correlation, the stronger the job market, fewer people decide to pursue an education because they're making too much outside. So I think it will be more that factor, and we'll see peak season. But I'm confident in our ability to continue gaining share.

For personal loans, we took a little longer to return to pre-pandemic credit criteria for that product, just given the higher volatility. But as John said, quarter-over-quarter, we're now flat. And so we would expect that to return to growth in 2022.

And the balance transfer activity, there's a bit of a link between personal and card. I know balance transfer comes before personal loan growth, but how's that legging in at this stage? Is that -- has there been a take-up beginning there yet? Or is that more of a back half '22 outlook?

So John talked about promotional balances as having an impact on NIM. I would say a lot of that is driven by new accounts. So as you ramp up new accounts, you'll see that, but also portfolio activity as well.

And we will take our next question from Kevin Barker with Piper Sandler.

Given your growth rates that you're projecting out there, I mean, do you feel like you can achieve a sub-40% efficiency ratio some time in the foreseeable future? Whether it be run rate close to late '22, maybe early 2023?

Yes. Thanks, Kevin. We specifically commented on positive operating leverage. So as we deliver that, obviously, the efficiency ratio will improve. My expectation is that we can get into the high 30s within the medium term. So we built the plan essentially contemplating significant investment in marketing and then working through the other elements of the cost structure in order to create as much efficiency and capacity to drive new growth. As that model continues to build upon itself, my expectation is that those high 30 numbers are certainly very, very achievable.

And then a follow-up on some of your comments around credit. I believe you said you expect it to increase sequentially throughout the year. Did I hear that correctly?

And then I mean, could you just give us a little bit more detail on your expectations for the cadence of net charge-offs, given we're at an exceptionally low level? And typically, you have quite a bit of seasonality. Can you just give us a little bit more color around your expected cadence on charge-offs throughout the year?

Yes. So it will be difficult to expand more deeply upon the comments that I've already made. So we do expect it to increase sequentially. There is some degree of seasonality, I don't view that as like a material driver to what we're going to be seeing. And I also mentioned that the charge-offs are more weighted to the second half of the year than the first half of the year, and that we've got pretty decent line of sight to the first half. So that's split out somewhere around 45% first half, 55% second half, is probably as deep as I can go on the charge-off numbers right now.

And we will take our next question from Don Fandetti with Wells Fargo.

Roger, as you come out of the pandemic, I was just curious if there's any areas of strategic interest, new products, et cetera, that you're looking at. It's been pretty consistent for the last several years.

Yes. Certainly, we're always looking for new opportunities on the payment side of our business. And there, the versatility of our capabilities, you saw some of that with our partnership with Sezzle in terms of our ability to provide easier connectivity to merchants. So on the payment side, both in the U.S. and globally, we're looking for opportunities.

On the card side, we feel really good about the product set we have. We are virtually 100% focused on consumer. I think there is a huge opportunity to continue to grow our non-card products. And so we talked about investing more on marketing our deposit products before the pandemic. Clearly, when we were in a significant excess liquidity position, it didn't make sense to put a lot of marketing behind deposits. But that's something I would expect to see in 2022.

Okay. And any changes on your international acceptance push? Or is it sort of you just kind of inch your way into growth?

We continue to push out. I think what you heard in the call, our favorite way of expanding internationally is through network-to-network partnerships. It's just much more cost-effective than working with individual acquirers, although we do that as well.

There's also a big focus on acceptance in the U.S. around the migration to digital. So working with other networks on secure remote commerce, everything from transit implementations. But again, we're now up to 26 net-to-net partnerships, and feel that there's room to continue growing. Recently announced a partnership in Serbia actually this week.

And we will take our next question from Dominick Gabriele with Oppenheimer.

I just want to follow up on one of the answers you had before. Do you think that the ability to reach more customers and spur spend through marketing has an overall direct relationship between which FICO band or income level these customers have?

And is this why you think perhaps there could be those diminishing returns on marketing investment at perhaps slightly different overall rates between what customer base one issuer may have versus another? And I just have a follow-up.

Yes. I don't think we said we expected diminishing returns on marketing. Credit is pretty similar to what it was pre-pandemic and I mentioned our projected cost per account. And a lot of our marketing does go to new accounts. Cost per account, we're projecting that to be pretty much on top of where it was pre-pandemic.

Different issuers certainly have different business models. There's one who's particularly focused on subprime, others are much more aggressive at the super-prime. We have been very clear for many, many years that ours is a lend-focused business, going after that prime revolver segment. We've tailored our products for that, our underwriting capabilities. And again, we feel very good about the return we're getting on the dollars we spend in marketing.

And can you maybe talk about how the competitive landscape is evolving? And what products are likely to either meaningfully compete, not ultimately compete and somewhat compete with your everyday spend credit card products? And perhaps maybe which spend categories may be most of that competition could reside, versus least likely.

Yes. I think clearly, what everyone's watching is buy now, pay later. As I mentioned earlier, we haven't seen that have a noticeable impact on our base. In my mind, it's closer to traditional sales finance. So there have always been competing products out there, whether it's on the private label side, whether it's personal loans for debt consolidation, et cetera.

So we focus on getting a broad mix of spend. Even through this year, we're seeing strong performance across every category. Travel is holding up, actually in January, surprisingly well given the state of that pandemic. So we think we'll continue to grow across categories. And in fact, as one of the beauties of our 5% program, it sort of reinforces different categories of spend on a rotating basis as opposed to products that are really particularly tailored to an individual category of spend.

And we will take our next question from Robert Napoli with William Blair.

Roger, John, nice quarter. I really like the guide on loan growth versus expense growth. Just on the expense growth side. I mean, if you -- there's a lot of investment going on at -- certainly at some of the major banks on technology. And I know that Discover has invested in technology over the years. But I would just -- maybe just your thought process on where your tech stack stands. Is -- your thoughts on private cloud versus public cloud, and the need to invest to compete over the next several years.

Yes, good question. I think our perspective may be a little different than some of the big banks. And of course, we have a narrower range of businesses. So I can't really comment on some of their investment. Certainly, we are focused on competing with the fintechs, but it's not just about spending money. It reminds me of -- it's like someone saying, "I'm just going to keep eating more and more until I lose weight." The competition -- those fintechs are not spending more on technology. Our focus is around capabilities, it's on agility, it's on speed to market.

And so if you look at last year, once you sort of sort through for onetime items, technology spend was relatively flat, but that doesn't mean there wasn't a huge focus around our capabilities. We've talked about our investments in data and analytics. So it's really more about speed, and you don't get there just through sheer dollars of spending.

Okay. And I guess -- I mean, I guess private cloud versus public cloud? And the importance of your cloud strategy. [About innovation...]

Great question. We're focused on a hybrid cloud strategy, so a mix of both. I think there's -- sometimes companies seem to take a purist element, that there's something great about having 100% of your applications on the cloud, whether or not you're GL resides on the cloud is not going to really make a difference for your business. But we are heavy users of the public cloud, in particular, for our data and analytics areas where the speed and massive amounts of storage are critically important.

And if I could just sneak in, your spend growth was really strong. How much of that is inflation? But what are your thoughts on inflation, how it affects your business? And the impact that maybe it's having on the spend growth numbers you reported, which were pretty strong.

Yes, Bob, I'll take that. Yes. We've been really pleased with the sales running through the card. The inflation has had a small impact. So you think, kind of on average, maybe 1% to 2% in '21. In '22, we didn't model that out specifically, my sense is it would be 2%.

We will take our next question from Mihir Bhatia with Bank of America.

Maybe first, I just wanted to go back and just clarify a little bit about your guidance. I just wanted to make sure I understand some of the key assumptions embedded in there. So first, like on payment rates, I think you said you expect it to slightly decline a little bit in 2022.

And I was wondering, if we were to see normalization happen a little bit faster, let's say they normalize back to the normalized levels in 2022, would that push your loan growth up to like low double digits? Or am I -- or is that like too ambitious in terms of like the impact? What I'm trying to understand is the impact of the payment rate on loan growth there.

And then just also related to just this guidance question just on rate hikes, and I apologize if I missed this in your earlier comments. Did you say how many rate hikes or how many basis points you were assuming in your guidance?

Yes. So we assumed -- thanks for the questions. We assumed 2 rate hikes in our guidance. And in terms of payment rate, as I said, very modest improvement, so -- or a reduction. So if payment rate were to normalize at the pace you just described, which I don't think it will, but if it were to do that, it would certainly be very accretive to the loan growth. I'm not going to get into specifics whether it takes us to double digits or not.

Okay, no, I understand. That's helpful, though. And then just wanted to -- the other question I will ask is just about understanding your credit underwriting or risk appetite currently. You did mention normalization is happening maybe a little bit more on the lower FICOs. Have you tweaked your underwriting or marketing in the last few weeks or quarters in response to that? Or is it still very much as expected, so it's all systems go?

Yes. No, we haven't. And I think it's important to note that the life of loan losses assumptions we use for new account underwriting is not driven by where losses are currently. So John -- [all] portfolio performance and what we're seeing. But we use sort of by segment and actually by individual account forecasts of life of loan as we determine our marketing and credit criteria.

And we will take our next question from Bill Carcache with Wolfe Research.

Is there any reason to be concerned about the pace of credit normalizing faster than receivables growth?

From my perspective, no. And I'll tell you why. First, the macros are -- the macroeconomic conditions that are super-positive, right? So there's more job openings than there are people looking for jobs. The consumers will have more dollars into their paycheck as a result of this inflation. And there is a substantial amount of savings still left in the economy from -- largely from kind of change in behaviors and government stimulus. So I'm not frankly seeing that as a risk in 2023 -- excuse me, 2022. As we get out into 2023, there's less certainty around that.

Maybe a related follow-up. There's a lot of consternation around NCOs normalizing higher. But do you think that the environment that we're in today, where you're seeing the revenue benefits from the accelerating loan growth that you're putting up on one hand, and then also on top of that, you've also got the reserve rate remaining well above day 1 levels. And so when you think about the risk of growth math headwinds and the need to build reserves on that strong loan growth that you're seeing, the risk that, that overwhelms the revenue benefits from the loan growth and the fact that the reserve rate is higher, can you talk to that interplay and how you're thinking about that?

Yes. So when we -- let me start with the reserves because it plays into the rest of your question. So when we looked at reserves for the quarter, what we wanted to see was the impact of the ending, the end of most of the government support programs. Most ended in September. So we had 1 quarter worth of data. We looked at it, we saw no real change to the portfolio dynamics. We're hoping to see another quarter and then reevaluate overall reserve rate. But that day 1 number and the -- I'll say, the normalized charge-offs in the 3s would support a view that, down the road, we'll have some opportunity on reserves.

In terms of new vintages and the kind of charge-off impacts from those, this company has been through years and years of cycles with new vintages. We're very thoughtful in terms of how we do the underwriting. We've got some improvements from the -- within underwriting from the advanced analytic tools that we've put in place. So my sense is as the portfolio seasons, we're going to see some increase in charge-offs, but well within the expectations of how we underwrite and well within the expectations of this guidance and the macroeconomic outlook.

And we will take our next question from Meng Jiao with Deutsche Bank.

I wanted to follow up with the question on the competitive environment. We've seen a few competitors coming in with new offerings. And I was sort of hoping for some more color as to whether you've seen any read-throughs to possibly yields, possibly tightening, or sort of anything else outside of elevated marketing spend with this increase in competition.

Yes, great question. In general, the competition takes the form of higher rewards, increased marketing spend. You'll see some players start putting up big onetime signing bonuses as they look to grow. Some players may start extending their promotional periods. But given sort of the inability to reprice cards post CARD Act, you tend not to see it drive yield compression.

And as I mentioned earlier, a lot of those products in the competition seems to be targeting that sort of super-prime transactor segment, that is not one that we aggressively go after. So our projection for flat cost per account next year reflects our view on our ability to compete in this environment.

And then secondly, John, I wanted to sort of circle back in your comments regarding the difference that you're seeing between higher and lower FICO scores. I mean, are you sort of seeing that in early-stage delinquency sort of diverging? Just any other further details that you might be able to provide in that specific difference.

Yes. It's in the kind of roll to one, so on delinquent bucket out there. And the fact that we're talking about it is, I'll say, is intended to indicate that we're paying attention to the entire portfolio, and that we have a growing comfort in terms of how the credit outlook is in the performance of the company into '22.

And we will take our next question from Bill Ryan with Seaport Research Partners.

Just a couple of things. First on promotional balances. Looking at your portfolio today as a percent of the total, where are you versus kind of like the history of the company or the historic norm, if you will? And what is the typical duration of the promotional balances?

And then I'll go ahead and ask the second question. But looked like there's a little bit of a drop in protection product revenue this quarter as a percent of the portfolio. I was just curious if there's any specific call-outs there.

Okay, great. So the kind of the promotional balance content of the portfolio, we tend not to kind of go into expansive detail about that. So my comments on net interest margins should give an indication that we intend to use that as a tool to help some origination activity.

In terms of the protection revenue, so we have an existing product that we stopped marketing some time ago. It's essentially providing value to the customer set. But if you don't market a tool, obviously, your revenue line gets impacted. We have a new product that we launched, very, very soft launch that we actually haven't done any broad marketing yet. So my expectation is that, that line will be flat to down in '22.

All right. Well, Britney, I think we're going to conclude our call here, but thank you all for joining us. And if there's any additional follow-ups, please reach out to us here in Investor Relations. Thank you, and have a great morning.

This does conclude today's program. Thank you for your participation. You may disconnect at any time, and have a wonderful day.