Bread Financial Holdings, Inc. (BFH) Q3 2017 Earnings Report, Transcript and Summary

Good morning, and welcome to the Alliance Data Third Quarter 2017 Earnings Conference Call. At this time, all parties have been placed in a listen-only-mode. Following today's presentation, the floor will be open for your questions. [Operator Instructions] In order to view the company's presentation on their website, please remember to turn off the pop-up blocker on your computer. It is now my pleasure to introduce your host, Mr. Edward Sebor of FTI Consulting. Sir, the floor is yours.

Thank you, operator. By now you should have received a copy of the Company's third quarter 2017 earnings release. If you haven't, please call FTI Consulting at 212-850-5721. On the call today, we have Ed Heffernan, President and Chief Executive Officer of Alliance Data; and Charles Horn, Chief Financial Officer of Alliance Data. Before we begin, I would like to remind you that some of the comments made on today's call and some of the responses to your questions may contain forward-looking statements. These statements are subject to the risks and the uncertainties described in the Company's earnings release and other filings with the SEC. Alliance Data has no obligation to update the information presented on the call. Also, on today's call, our speakers will reference certain non-GAAP financial measures, which we believe will provide useful information for investors. Reconciliation of those measures to GAAP will be posted on the Investor Relations website at www.alliancedata.com. With that, I'd like to turn the call over to Ed Heffernan. Ed?

Great. Thanks Eddie and welcome everyone. As usual, joining me today is Charles Horn, our CFO. And he's going to kick it off with an update on third quarter results, and then I'll jump in with full year expectations, and then provide the next layer of detail to our 2018 guidance. Charles?

Thanks Ed. Revenue for the third quarter increased 1% to $1.9 billion, soft against our mid-single digit expectations, primarily due to hurricanes Harvey and Irma negatively impacting the growth shields at Card Services, which knocked off about three points of growth. Core EPS increased 13% to $5.35 for the third quarter 2017, better than our expectations, primarily driven by lower loss rates of Card Services and a lower tax rate, a result of initiatives implemented during 2017. Excluding the improvement and the effective tax rate, core EPS increased approximately 9% for the third quarter. Adjusted EBITDA net increased 7%, benefiting from the better-than-expected loss rate at Card Services. During the quarter, we reduced our corporate debt by approximately $180 million to $6.2 billion. That brings our corporate leverage ratio down to about 2.8x as of September 30, 2017 versus our covenant of 3.5x, leaving us with over $2 billion in available liquidity. Let's slip to the next slide and talk about LoyaltyOne. As expected, the third quarter was tough for LoyaltyOne, with revenue decreasing 20% to $305 million and adjusted EBITDA decreasing 26% to $61 million. Fortunately, the future looks brighter for this segment. Breaking it down by major business. AIR MILES revenue decreased 28% to $185 million for the third quarter of 2017, driven by a 43% decline in AIR MILES reward miles redeemed. This drop in revenue was consistent with our expectations as we pulled in the burn rate, which are miles redeemed divided by miles issued from 129% last year to 80% this year. 80% is our target, really, moving forward. Importantly, the negative draw over from the elevated redemption rate is gone entering the fourth quarter of 2017. AIR MILES issued was down 7% year-over-year. Normal spend is on track, but we have seen less promotional activity in the grocery vertical. We expect issuance growth to improve in the fourth quarter and in 2018. Importantly, we recently re-signed BMO, our largest AIR MILES sponsor, to a multi-year contract. BrandLoyalty's revenue decreased 5% to $121 million for the third quarter of 2017, consistent with our expectations. BrandLoyalty's fourth quarter is looking good, with revenue and adjusted EBITDA expected to increase double digits year-over-year. Let's go to Slide 5 and talk about Epsilon. Epsilon came in short of expectations for the third quarter, with revenue increasing 3% to $559 million and adjusted EBITDA decreasing 7% to $125 million. Year-to-date, revenues has increased 5%, slightly better than annual guidance while adjusted EBITDA is essentially flat, soft compared with our annual guidance of 4% growth. Revenue came in short of expectations as the growth rate in our digital offering slowed during the quarter. We believe the onboarding of new clients will fill this hole quickly. The pull back in digital growth also negatively impacted EBITDA margins for the third quarter. Importantly, Epsilon's technology platform offering continued to show progress toward a promising turnaround, with revenue increasing 1% for the third quarter. This is a significant improvement compared to the double-digit decline in the fourth quarter 2016. The client base is showing stability, and we've seen an increased win rate on the basis of the introduction of cloud-based more packaged products into the market. Let’s now move on to Card Services. Card Services produced another solid quarter growth, with revenue up 9% to $1.1 billion and adjusted EBITDA net up an even better 20%. Average card receivables increased 14%, just under $16 billion, consistent with our expectation of mid-teens growth while credit sales increased 5% to $7.4 billion. Operating expenses increased 16% to $382 million or 9.2% expressed as a percentage of average of receivables. The percentage is essentially being flat with last year. The net loss rate was 5.5% for the third quarter 2017. A substantial improvement compared to the 6.2% loss rate in the second quarter of 2017 and better than our expectations. Importantly, gross loss rates were flat with the third quarter of 2016, even excluding the slight benefit from the hurricanes, which I'll talk about later. Recovery rates are still down year-over-year, pressuring the net loss rate, but we consider this pressure to be transitory as we shift collections back to internal. Our delinquency forecast continues to suggest flat to lower loss rates in 2018. This trend allowed us to slightly drop our reserve rate during the quarter while maintaining 12 months of forward coverage. Now the question everyone has been asking what was the impact of the hurricanes. We estimate the hurricanes reduced our credit sales by approximately two points, reduced our revenue by about $40 million for the quarter or about 100 basis points negative impact to gross yield and benefited our loss rates by about 15 basis points. So what that means to our gross loss rates were down 18 bps year-over-year, take away the 15 basis points for the improved from the hurricanes, and that's roughly the flat year-over-year change in our gross loss rates. It is important to remember that this is a temporary impact. Under FEMA-designated individual assistant disaster areas, affected cardholders are not required to make payments nor will any interest or late fees be assessed for two month period. Hurricane Irma will impact us early in the fourth quarter and then operations will return to normal. With that, I will turn it over to Ed.

Great. Thanks Charles. Slide 9, full year 2017 outlook on a consolidated basis, we're going to reiterate guidance of $7.8 billion in revs and $18.10 in core EPS. Obviously, on the rev side, the hurricanes hit us, as Charles said for about $40 million in Q3. We expect another $40 million in Q4. So it is about a $80 million on the top. Despite that, we are going to maintain the $7.8 billion guidance for the year. In terms of earnings per share, despite the big beat in this quarter, we are keeping it the same for now. If you flow through the impact of the hurricanes all the way down, we expect that to be about a $0.25 headwind between Q3 and Q4. We think we can play through that primarily because – as Charles mentioned, our loss rates are actually coming in a little bit better than anticipated even after the impact of the hurricane. So for now, we're going to use that as sort of our cushion to make sure we can play to the hurricane and hit the $18.10. So what does it all mean? It means that the slingshot that we've been talking about it seems forever is finally on track. The delinquency wedge is closing this month, which we're all looking forward to, and we expect Q4 to have our delinquencies finally anniversaried, and that means good news for next year. As Charles also mentioned, growth losses, which were up 75 basis points in the first half of the year, we thought they'd be up about 30 in the third quarter and flat in the fourth quarter, actually came in slightly better in the third quarter, and were, in fact, flat in the third quarter versus prior year even after factoring out the hurricane. So we're running about a quarter earlier in terms of the benefits of the lower loss rates, which is, obviously, good news and led to the overperformance in the quarter. Recoveries also remain soft due to market conditions, but should they make 2018 a pretty easy comp from that perspective. So, you put it all together and slingshot looks like it's shaping up pretty nicely. The momentum is building as you see our core EPS went from 2% in Q1 to 4% in Q2 double digits Q3 and we're going to be high single, low double in Q4. So it's getting ready for a nice run here in the back half. And as you move sequentially from Q3 to Q4, the big mover in Q4 finally after three sort of soft quarters, will be LoyaltyOne, which sequentially will add $0.50 a share more than it did in the third quarter. So that's the big mover that we're looking for. And let's turn now to Slide 10, which is the 2017 outlook. We start with Epsilon. The original guidance was $2.24 billion in revenue and about $0.5 billion in adjusted EBITDA, 4% top and bottom. Year-to-date, we're running a little bit ahead on revs and a little bit behind on EBITDA. But in general, we're right in that range where we needed to be. On the Technology Platform side, which was, if you recall, the big issue last year, it was sort of a melting iceberg there for a while. Revenues were actually down 13% in Q4 of last year, then it was minus 7% and minus 3% and actually this quarter, it turned to plus 1%. And so we think for now, we've got a pretty good miles traffic there that's going to be additive next year. All right, let's go to Slide 11, which is LoyaltyOne, which has the two components, Canada and then BrandLoyalty or the business based over in Europe. Canada, no change in guidance. From beginning of the year, we expect a little over [$750 million] in revs and $180 million in adjusted EBITDA. Looks like we're coming in fine with both of those. We're making solid progress on retooling the model, the margins are running in the mid-20% range, which was what we were shooting for. Probably the two biggest pieces we were focused on this year would be the sponsors, those are the folks that pay us. We've seen no attrition from the sponsored ranks, and as Charles mentioned, we just renewed our biggest one, the Bank of Montreal. So that bodes well. On the collectors' side, the question was are they going to be reengaged? Are we going to lose a bunch of active collectors because of the noise last year? And we were down about 6% in the latter part of 2016 in terms of total numbers of collectors who are active in the program. That's now to within 1% pre-crisis, as we would call it. So looks like the collectors are hanging in there. The one are that we're still working on while the core issuance is stable, which is about two thirds of all of the issuance, the promotional-related activity is still soft, and we want to start seeing that from up as soon as Q4. So, we've got a couple of ideas on that down that front of a couple of big programs we're launching. BrandLoyalty, as we talked about last quarter, was probably the surprise this year in the sense that it was soft through all three quarters of the year and traditionally and historically has grown double-digit top and bottom. You could pretty much set your watch to it. And as we looked into it, we wanted to make sure that there wasn't a structural issue, that was more of a timing or transitory issue, and it looks like we're going to be in good shape for Q4. We expect return to strong double-digit in Q4 and give us a nice jump-off for next year. So it was bit painful for the first three quarters and then, we're finally going to see some of the benefits in Q4. All right, Card Services, where we get about 99.9% of our questions. Let's talk a little bit about that. Card receivables growth of 15%. You will see that on some of the monthly reports, we get some questions of, oh, it looked like it decelerated a little bit from earlier in the year, and is that the trend? And the answer is, no. What we did is we proactively went out there and there are a handful of portfolios that are viewed as noncore at this point that have been moved to held for sale and – since they no longer fit. It also clears the deck for us to bring on Signet without really stretching ourselves and keeping us that 15% growth rate for the year. So that should be where we wind up for the year. We signed $2 billion vintage, and you'll see a handful of announcements coming before year-end that will be additive to what we have announced so far. Gross yields, as Charles talked about are down in the third quarter and will be down in the fourth quarter due to the hurricane impacts and the onboarding of Signet, and we can get into that in Q&A. We are seeing some decent operating leverage of roughly 40 bps on operating expenses that excludes losses. And then the big question, credit losses, credit normalization. Are we there yet? Is it almost done? Bankcard players seemed to be saying that it doesn't look like they're going to normalize to the latter part of next year. And where we're sitting right now is a little bit actually ahead of where we thought we'd be. Delinquencies are right on track with where we want them to be. They're the best predictor of future gross losses. Q1, we're about 50 basis points over last year. We are on track in Q4 to be flat versus last year. There was a little bit of noise last month due to the hurricane, but we had sort of our final bubble of accounts flow through. And so we expect Q4 to be flat, and that's what we've been shooting for all year. And so from a gross loss rate perspective, Q1 and Q2, we were up 75 basis points versus last year, as I mentioned, we thought we'd be up about 30 or so this quarter and then finally get to flat in the fourth quarter. In fact, we – gross losses were actually slightly better than they were a year ago in Q3. If you factor out a slight benefit from the hurricane, we were still flat to where they were last year, which, again, is running about a quarter in front of what we had anticipated, which is great news. So I would say that the outlook on the credit front, I would say, is looking a little bit better than we had anticipated, which again reaffirms the impact of the slingshot for next year. Then we get to net loss rates, again, for those of you who don't live in this area, you have gross and then you have recoveries against that that will give you net. We talked about recoveries moving from higher than a year ago to flat. And now we need our recoveries to move from sort of lower than they were a year ago to actually flat. And so that's going to happen as we move recoveries in-house and should provide us a pretty easy comps for next year on that. So what's interesting on the net loss rates is when we started the year, we guided to approximately a 50 basis point increase in net loss rates. And if you look at the components, we assumed it was 50 basis points on gross losses and recoveries would be relatively stable as they have been for years. It turns out the gross loss rates are only going to be up about 30 basis points. So we're doing better from that perspective, but it's certainly overwhelmed by the fact that the recovery market fell apart at the beginning of the year and so most of the increase in our net loss is due to the recoveries falling off. So it's a mouthful, but what we focus on is what that say about going into next year. And right now, we look pretty encouraged. So loss rates 6.3% in Q1, 6.2% in Q2, 5.5% in Q3. It will be below 6% in Q4. It will flow something like 6%, 6.5% for October, November, December, something like that. Full year guidance, we actually expect mid-teens revenue growth in cards, and we expect 12% growth in adjusted EBITDA, which again is net of all the funding cost in cards as well as the provision for credit losses so double-double for cards despite going through this normalization process. All right, let's go to 2018 guidance. We said we're holding on 2017 at $7.8 billion and core EPS of $18.10. For 2018, we're also holding at our $8.7 billion, up 12%, and $21.50, up 19%, a significant jump from where we are this year. And is it reasonable? First thing we look at is what our10-year average has been, and we've been running about 13 tops, 17 bottom. So it seems like after a softer year like this year that coming back stronger next year will kind of makes sense with the overall model. So that's what we're shooting for. If you go to Slide 14, let's give you sort of a little bit more detail on the businesses. Epsilon, we expect mid-single-digit revenue. In EBITDA growth, we expect a big Technology Platform business, which builds the big CRM databases and the loyalty programs to return to modest growth, which is from our perspective very good news. And the Conversant CRM and auto verticals remain our fastest-growing offering. So mid-single, mid-single seems to be in order for Epsilon. LoyaltyOne, this is the year to fully retool the program. We think it has been retooled. We want to see 3% to 5% growth in both top line and EBITDA. We want this business to return to modest, consistent growth and we believe that's certainly doable. There's a lot more wood to chop here in terms of what this business can do in Canada. And at a minimum, we would expected it to grow at the nominal GDP growth rate of Canada, which is a little over 3%, get a little leverage and perhaps we can get to 5%. On BrandLoyalty, as we’ve talked about, we expect Q4 to be the start of a nice run after multiple years of double-digit top and bottom, this year. Obviously, soft, we expect that to come back in Q4 we have pretty good visibility on Q4 and give us a nice jump-off for Q1 of next year. The comfort level that we're going to have for 2018 is really around the fact that in years where there's some big global event such as the World Cup, we tend to have more activity and more participation. And so next year, we have one of those events. We also have that big deal signed with Disney, and we'll be using that from a reward perspective across Europe. So I think it looks like the pipeline is shaping up pretty nicely. Cards, where is cards going to be? Cards is going to be mid-teens portfolio growth, supported by this year's $2 billion vintage that we signed. We expect a nice solid return to mid-teens growth in both revenue and adjusted EBITDA. Again, it's a funny term, but adjusted EBITDA is net of all the funding cost and the credit cost. So and we expect net loss rates to be flat to down, meaning flat to improved versus 2017, meaning that the full normalization has now taken place and it should be very stable from now on. Okay. I want to finish with 2018 guidance. I specifically want to address where probably the vast bulk of questions have been coming in over the last year and a half or so, which is loss rates and what does it mean, and how do you know it's going to be better and all that other stuff. So we've tried to choose – I tried to sort of put it in a one page or so to speak. And what you see with Private Label and having lived through the Great Recession and a couple before that, what you'll find with Private Label is their – the loss rates in Private Label tend to peak in front of the peak of loss rates for bank cards. Said differently, we started talking about our normalization sometime back in late 2015, and we said that 2016 and 2017 was going to be the two years that it would take to bring losses back up to sort of pre-great recession levels after being abnormally low. You started to hear from the big banks and bank card players about a year after that. And so the way it tends to work is our normalization runs about a year in front of the bank cards. And that's playing out exactly the way we've laid it out here, which is we're going to be done with ours in the Q4, whereas, I think the big bank card players are probably indicating its going to take them to late 2018 to normalize. So we started earlier, we'll finish earlier and that seems to be the trend that it's following here. As a result, the huge reserve builds that we put through in the last couple of years are behind us. And now, you're looking at more of it builds with the growth in the file itself. The second reason why we feel we're in good shape in 2018 is the delinquency wedge, which again our accounts that are flowing through being late on payments for 180 days before they're written off. That gives us a good view into what's coming over the following six months. The so-called wedge, where delinquencies were up year-over-year is closing in Q4, and we expect it to be flat to prior year in Q4 itself, which means going into 2018, that’s a good signal that losses will be flat as well. Next, to sort of pile on here with Signet, which is coming on latter part of this month, talking about $1 billion plus portfolio, where if you recall, we only purchased the prime-only accounts. And so that carries with it slightly lower overall loss profile, which, again, can help dampen your losses on a go-forward basis. And then finally on the recovery rates, we were running this year at about 18% of gross losses. And normally, we recover somewhere in the 20s, and that had a lot to do with the third-party collectors out there and some new regulations and a lot of paper on the market. We made the decision to move the business over time back in-house, which is what we've done before. We did it post-Great Recession. And it's really just math of how much we can get on the outside versus how much it cost to step up and do it inside. We know we can do better than 18%, somewhere in the low-20s is where we'd like to be. So 2017 was unusually low. So moving in-house will gradually increase the recovery rates, basically saying, you get some decent comps as we move into next year. So everything is pointing to the fact that loss rates should be stable, to improved next year. It's been a very tough couple of years for all involved, including those listening on the phone. And to give you a sense of what normalization cost, it's fairly punitive. The higher loss rates and reserve builds were the double-digit drag on core EPS, really, in both 2016 and 2017. And to put it in perspective, in 2017, with loss rates up 80 basis points, 30 gross and 50 on recoveries, we have core EPS of $18.10, up 7%. Had loss rates just stayed flat, and we set the reserves and provisions just based on portfolio growth. Our core EPS would have been up over $21, so up 20%. That gives you a sense of how important it is that we reach the end of this normalization process. That's about $260 million of EBITDA that was poured into this expense over and above flat loss rates. So if you looked over the past two years, 2016, 2017, this normalization process has cost about $400 million of EBITDA or pretax earnings due to both the higher losses as well as the higher reserves we had to set aside. We think that with the flat gross losses and the flat delinquency wedge in Q4 and relatively easy comps and recoveries. We'll be flat to down next year on in terms of losses, and that should very simply math return us to the high teens on the core EPS. And so after two years, the big reserve builds are over. And we've got very good visibility – actually excellent visibility on where we're heading in 2018. So I think we've feed now on the depths. So why don't we go ahead and move to Q&A?

Thank you. [Operator Instructions] Our first question comes from the line of Dan Perlin of RBC Capital Markets.

Thanks, good morning, guys. So I had a question around card, in particular, around the operating expense. So – and it relates to where you are in the build-out of your in-house recoveries. So was up, I guess, 16% or so year-on-year, but as a percentage of AR, it was only modestly up year-on-year. It was really was up sequentially, but modestly. So I'm trying to understand how we should be mapping that out over the next several quarters. And where are you in the process of actually bringing those people and to build up the in-house recovery?

Yes, so there's two things I'd point you, Dan. First is the gross yield decline year-over-year makes the percent growth look a bit higher, the 16% you referred to. If you had basically the same gross yields year-over-year, you'd see it flat-flat. As a percentage of AR, it moved a nominal amount. We said, going back to the last quarter, we expected our operating expenses to be better this year, 40 to 50 basis points. I think that's still on track. We would tell you we're still in the process of adding collectors. And we'll staff them up over the remaining of Q4. I would still expect the OpEx leveraging in Q4 to be better year-over-year. That said, we're still very much on trend to that guidance we gave last quarter of about 40 to 50 basis points improvement on an annual basis.

Okay. But just to be clear, the absolute dollar of OpEx expense was up 16%. And you're saying you're expecting that to stay at these levels or playing out throughout the – I would imagine that's where you're capturing the incremental cost of this recovery business, correct?

That'd be correct as well. But let's say we hadn't had the influence of the hurricanes affecting our gross yields, the revenue growth would have been up 15% to 16%. So you have to basically influence the fact that you take a whack on your gross yields is not influencing your OpEx.

Okay. The other question I had is, can you just help me understand why the 15 basis point improvement in gross losses as a result of the hurricanes? I mean, I think I caught it, but I'm sure someone will ask, and I just want to make sure I'm clear on that.

It just means for affected person within that FEMA designated zone, that account is frozen from migration to the delinquency trends to loss if they're in that affected area. So if they were in to 150 days past due and they were affected, it would just mean that's going to freeze and that would not have rolled into the charge off after 180 days.

Okay.

Yes. So essentially, you get hit pretty hard – or actually harder on your lack of earnings in terms of the revenue, but you are getting a slight benefit in terms of – for a couple of months of putting those accounts – or suspending those accounts from rolling to losses. The net result is going to be about a $0.25 hit Q3, Q4 combined.

Yes. Okay. And then if I could sneak in one just quickly on Epsilon. You mentioned a pullback in digital growth. I'm wondering what do you think is driving that for you guys. And then secondly, you said onboarding new clients, I think, is going to help drive a big part of reacceleration there. And I'm just wondering what's happening, really, in the underlying core business. And then why is this – why are we seeing this shift? Thanks.

I'd say, Dan, we think the core is still very strong. Occasionally, you get some timing differentials when new programs are coming onboard. I think that's the biggest thing we're looking at. We'd expect it to bounce back in Q4. We just had a little bit of a timing with fewer new programs launched in Q3 than what we've been doing before. But like I said, I think that will bounce back in Q4.

Okay. And then the onboarding as sort of rate of change of new clients versus core, it's just proportionate to the new clients signed, I guess, is what you're saying in terms of how we're mapping up growth.

We would tell you the backlog is still strong. Occasionally, it just shifts as to which quarters till when the program is going to onboard. And I would say, we said a little bit lower in Q3.

Okay. Thank you, guys.

Thanks.

Our next question comes from the line of Sanjay Sakhrani of KBW.

Thanks, good morning. Charles and Ed, I guess, you guys mentioned the couple of different numbers on the impacts from the hurricanes. But I wanted to make sure I got the total because when I look at Slide 3, it talks about a three percentage point headwind related to the hurricanes. Is that the consolidated number, which would be, like, closer to $60 million on revenues?

That is the consolidated number. And again, it's rounded to three. It was about $40 million impact to card and a $40 million to total. It's about 2.6% so around…

Okay, I got it. All right, and I guess, my second question is a little bit higher level on Epsilon. It seems like – and it may be related to the question before, but just seems like that, that business unit is supposed to produce higher – high single-digit revenues. And that's what Ed sort of talked about in terms of multi-year view. I guess, when we look at all the variability in that business, I mean, is that still the case? Do you still think, Ed, that's it's a high single-digit top line grower? Or is there to be going some variability over the course of the next couple of years?

Yes, it's a fair question. I mean, I think the one thing we've learned over the last couple of years is this variability is certainly by quarter. The big 25% of Epsilon, that's a tech platform was really our big concern because that was, as we said, melting pretty quickly last year. That's fully turned and we expect that to contribute positively next year. And so I really do think the long-term growth rate, Sanjay, of Epsilon has to be up around that 7%. And it's a question of how do we get from the sort of 5%. We're running this year, year-to-date, up to the 7%. And hopefully, with the technology piece turning, that should give us a decent shot next year to get pretty close. But my goal, quite frankly, is that, that's going to be not really a mid-single-digit. That needs to be a high single-digit top line growth business.

I got just one more on credit quality because I figured I should ask this, and I appreciate all the color that you guys provided related to it. But I guess, with credit now sort of outperforming even relative to your expectation, the concern in the market has been there might be some contagion with the consumer, like there's something wrong with the consumers' health. It seems like given that credit is doing relative better than your expected range and this is a range that you provided a couple of years ago, the state of the consumer is relatively stable. That – would you concur with that view? Or do you feel like the health of the consumer is a little bit weaker than when you first gave that guidance?

No, no. I – there's no question in our minds. And we do, if not twice a week, at least, weekly calls with the folks on the front line doing the actual collections. And these are thousands of people and we ask, what are the trends? What are you hearing? Are you hearing any sort of hesitancy? Are you hearing some of the noise that's coming out? And it's been rock solid with – there's no issues out there. If anything, I'd say, things are actually a little bit better. And obviously, we're seeing it in our own loss rates. But no, we're not hearing anything. And it sort of goes back – it's playing out pretty much the way we said, as you mentioned, a couple of years ago that we're just normalizing back to where we were before the Great Recession and then it stops. And it's stable thereafter.

Right, thank you.

Our next question comes from the line of Darrin Peller of Barclays.

Thanks guys. Just a first question on the growth of receivables guidance. I know you're mid-teens for next year, including Signet. I guess, that's a more of a high single-digit organic growth rate. It seems like your – is it strategically you're managing to a mid-teens all in growth just to really balance the business now? It's a bigger picture question, Ed, around the mix of business between the card business versus the others. Or is it more that the organic growth of the purchase volume has become more of a mid- to high-single-digit grower, and so that's just the new world we live in now?

If I can say yes and yes, I think that would probably cover it. But I think it's sort of a combo platter. They earn off exactly what you just said, which is, we think, 15% growth for us is roughly a nice comfortable growth rate given the amount of business that's flowing in to the top of the funnel and to your point, factoring in, which we didn't have several years ago, some attrition on the bottom of the funnel as some of our longstanding clients are having difficulty making the transition into the new, whatever the new is. And so we're not getting quite the growth we got from the core of five years ago. But what we're getting is we're getting more coming in, in terms of people shifting the marketing dollars to the data-driven business, which is Private Label. So it kind of watches against each other. And I would say, 15% is a good number. Could it flex up to a couple of points higher than that if core retail comes back? For sure. But I think 15% is a safe bet and that's sort of our strategic plan for the next three to five years.

Okay. And so just to remind us, the e-com – or the digital element of your business, I know you've talked about it before, I mean, that's obviously one of the bigger drivers. What is the metrics you've provided around that, again, if you don't mind repeating?

Well, what we said on the last call, Darrin, is that the credit sales through online are basically outside of bricks-and-mortar over a third of our credit sales. That number can jump up to north of 40% in the fourth quarter as you get the holiday spend. That's really the metric we've been focusing on. It's the traditional bricks-and-mortar or about 85% in store, 15% online. And we have over a third of our credit sales online. So that's really the ability for us to capture the data, identify you, track you online and give you a relevant target it off and it gives us the ability to do that.

Yes. And I would say that we're also beginning to see, Darrin, the core retailers. We're working pretty hard with them to increase the ease of their – of the online experience as well as bringing on the wafers of the world, the overstocks of the world, sort of the pure e-commerce players. And you'll continue to see that shift. My guess is – from an online credit sales perspective, my guess is we'll be at 50% in a couple of years.

Okay. That's helpful, guys. So just a quick follow-up, Charles. The delinquency trends you guys are guiding towards for the fourth quarter and into 2018, what is the impact of Signet on that in terms of – in other words, organically, without that would – I mean, I think delinquencies are still improving year-over-year and the wedge is getting close, but what are – where would we have been, I guess, organically?

We'd put it this way, Darrin, is the improvement is not definitive on us onboarding Signet. I mean, we're looking basically – yes. If we're looking to Q4, we would tell you putting Signet on the side line, we feel good gross losses will be down year-over-year in Q4. And that's going to be driven by the natural improvement you're seeing on delinquencies in that wedge.

Okay, all right. That’s great to hear guys. Thank you.

Our next question comes from the line of Ramsey El-Assal of Jefferies.

Hi, guys. Thanks for taking my questions. Can you take a step back on the hurricane and just help us think through what percentage of your business is generated in those impacted geographies?

We estimate that it's up to 15% of our accounts were within the Florida affected area and the Houston and Texas affected area. That doesn't mean that all of them would have gone into this program, but it's up to 15% of our accounts were affected.

Okay. So yes, essentially what you have is you put them into various buckets. And FEMA provides us with the areas that are hit the hardest. And that's probably about a two month tight hit, so you get the tail end at Q3. You get the beginning of Q4. But by November, December, we expect things to look pretty normal across the board. And we've got check this thing with what happened during Katrina, obviously, very different, but that's sort of – it's the same blueprint.

And basically, there's no kind of intelligence behind in the sense they just give zip codes, and any account associated with that zip code gets the special treatment.

No.

Okay. I wanted to ask about the Equifax situation and whether that has – you anticipate any impact from that. It feels like not as many people are freezing their credits as maybe was anticipated initially, whether that would impact you at all regardless because you have a different underwriting procedure. Just wanted to get your comments on that.

Yes, it's a great question. And I think it's top of everyone's mind. For sure, you heard an awful lot of stuff, bad stuff that came out of it and one of the things had to do with how can we protect the consumer or the consumer can freeze their accounts, et cetera, et cetera, that's going to lead to a lot lower level of account openings, both online and within the bricks-and-mortar stores for us. And we watch it pretty closely, and to your point, what started off as a potentially significant item and dwindled pretty quickly. We estimate that it's probably, at most, it will affect new account openings by about 1%. I mean it’s we’re just not seeing it happen. And even in some cases where they may freeze at Equifax, they're not freezing at the other bureaus, which we also use. So it remains to be seen a year or two from now, whether everyone gets on the bandwagon. But right now, it's really a nonevent from that perspective.

Are there any dependencies on Equifax in terms of data you require for underwriting that have – you've had to find another solution? Or is it really just kind of business as usual?

For us, it's pretty much business as usual. We'll go to other folks as we need them. But in terms of data availability that we need, I mean, that – it's there, it's available, and it's business as usual. It doesn't change the fact that a lot of sensitive information has gotten out there. But from our business perspective, it really hasn't changed procedures.

Great, thanks so much for taking my questions.

Sure.

Our next question comes from the line of Andrew Jeffrey of SunTrust.

Hi, good morning guys, thanks for taking the question. I guess a couple questions about loyalty. First would be in BrandLoyalty. Just trying to understand, sort of similar to Epsilon, how do you think about sort of your installed base of customers in terms of their growth. You mentioned some new signings and initiatives next year that will probably drive growth. But can you talk a little bit about same-store sales in BrandLoyalty, if that's the right way to think about the business and what those trends look like?

Yeah, it’s a good question. The way, it's a little different from same-store, but kind of the similar concept, which is the number of programs that a specific chain will launch in any given year. See you may have a large grocer in Germany that if we can get them to do two programs a year, that would be great, right? We're running 220, 230 programs a year, so it is a bunch. And so we make our money, essentially, when whoever it may be decides to run one of these three month promo, heavy-promo programs. So it could be either an existing grocer pulls up to do three programs in a year or we get a new grocer who only wants to do one program a year. So it's really hard to do a same-store sales. But in general, we know sort of the total number of programs we need to launch in a given year to drive what has historically been a double-digit top and bottom growth engine for us. What we're seeing, as we move into World Cup and everything next year, is the European side of the business, which is the bulk of it tends to crank it up quite a bit for these things, less so in North America. And that is, in fact, what we're hearing from some of our big clients over Germany, in France, in Italy, et cetera. We also – with the launch, we're always looking for new product for these promotions. And having the exclusivity with the Disney brands over there, I think, will provide an additional incentive for the grocers to roll it up. So there's really two drivers that we think will help the return to double-digit growth next year. And then the big wild card, Andrew, we're not baking it in, but if we can get someone big in the U.S. onboard, that's your upside. And so we've got a couple small programs that have launched in the U.S. Can we get one of the big guys to sign up, to try guys to sign up, to try to differentiate themselves a little bit, especially given all the turmoil in the grocer market here in the States? Don't know yet.

Got you. That's helpful. And then just as a follow-up on AIR MILES. There's been clearly some disruption in the U.S. grocery market. I know you talked about promotions in Canada and the desire – the intent to grow with GDP. Can you just comment on sort of the health and vibrancy of the grocery channel generally and whether or not that's something that you think is affecting AIR MILES?

Sure, yeah I think the reason for the drop-off in promo activity is, to your point, it's all grocer. And so we need to make sure that the other verticals are running and gunning, so that we can get this whole machine back up and going again. I mean, we're getting there. And it's frustrating in the sense of it hasn't fully turned yet. But the collector activation rates are pretty much back where they were. We haven't lost any sponsors. We're seeing weakness, to your point, on the promotional side from the grocers. But with the new deal with Bank of Montreal, which is a very large, while it is the largest sponsor, we expect to really reenergize that card program there. So hopefully, the game plan is we're going to get more out of the financial services vertical than we had in the past and that should help hedge against continued softness on the promo side.

Thank you.

Our next question comes from the line of Tim Willi of Wells Fargo.

Thank you and good morning. I have two questions around Private Label that, I guess, might be a little bit more intermediate term in nature. But just given the discussion around the operating leverage and the improvements that you've quantified, Charles, is there like sort of a permanent sort of target we should think about on the operating expenses relative to receivables on an ongoing basis once you sort of gets that peak, I guess, leverage point, that this is sort of where the ratio goes to and probably stays there if we keep it at a sort of a mid-teens portfolio growth? Anything we should think about?

Yeah. I’d hate to say anything is permanent, but I'd tell you what we would target is about 20 basis points improvement per year, and that's expressed as a percentage of average AR. Obviously, growth imbalances influences it. There are certain programs that influence it. But I think that's a very reasonable target. It's about 20 basis points per year.

My other question was about the digital side and referencing about 1/3 of credit sales and maybe, over time, that goes to half. Are you seeing anything around delinquencies or customer service related to the consumers that are heavy digital? Just sort of curious about shift over time impacts. Anything positive or negative for the operation on serving those retailers and their customers?

Tim, we'd tell you not really anything that we've seen different between the interactions online versus bricks-and-mortar. We know with online, we need to drive different service products. That's just the ability to minimize shopping cart abandonment, the ability to capture. You make it very simple if you make a transaction. But in terms of loss performance, delinquency performance, so forth, there's really no differences.

Yeah, I think, the thing to remember, this is sort to use the buzzword, the omni-channel approach of whether it's bricks or clicks. What you're basically looking at is the same customer. And five, six years ago, you would say that she would shop 8 to 10 trips a year. And those would be to the mall. Now she's shopping like a dozen times a year, some more trips, but only two to the mall and the other 10 are "trips online." So it's the same customer. It's just that they're shopping differently.

Great, that’s all I had. Thanks very much.

And we do one more.

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

Good morning, thank you for squeezing me in. One question on credit first and then an Epsilon question. Just the improvement in credit in the month of September, Charles, is that a move to the heavy increase in held-for-sale? I mean, there was a pretty significant increase in held-for-sale in the quarter. And when do you expect to sell those receivables? Are they essentially marked to where you expect to sell them at this point?

So think about it this way. It didn't influence our loss rates, really, at all for the quarter. The shift you saw that is put in held-for-sale is a vertical that's really just not a core vertical for us. It's pretty small. We're going to look to divest it. It is at the carrying value, which would be par amount than the anticipated losses, so there is no assumed gain or anything associated with it. It's something we'd probably look to try to get off our books toward the end of this year, early next year. But really, it goes back to something that Ed said before, is with the growth we're generating and now assume that coming onboard, we can look at different verticals and this is one we tested for several years, really, not a great vertical for us. It's really time for us to basically do a little bit of cathartic process to the file and get rid of the things that we don't want to focus on. And so that shift to held-for-sale, that's exactly that small vertical that we're just decided not to pursue.

And the improvement in September then was driven by – it was a monthly data.

It is performance of the overall portfolio, plus we did have some sales in the market, which we already had contracts in place. So just like we saw in June, we saw a little bit of improvement in recovery rate. We saw the same thing in September. So it's a combination of natural trends, a little bit of sale of previously charged-off paper that really drove the improvement.

Yes, gross losses themselves are actually running flat or better than last year, which, is about a quarter ahead of when we thought that would happen. And then you throw in some of the third-party sales on the paper that we had hanging out there from the beginning of the year. That's what drove it.

Okay and Signet, you said you think it's going to be a little over $1 billion now coming onboard.?

Yes.

Okay. And then just as it relates to Epsilon because that's – and I know you talked a bit about that. But the slowdown was obviously disappointing to you and below to us. But the CRM business and the auto – I mean, that business has been 30% grower. And it seems like having spent time with the management team there that seemed very optimistic about high levels of growth for that business. Is the visibility not as high? Is it not as predictable? And how lumpy – and are the opportunities in that very high-margin business, do you expect to see that? I mean, sounds like you're talking down the expected growth of that piece of the business.

We break it down it two ways. We'd say auto is still doing fine. That's really not showing the revenue growth. We did see the CRM drop. The digital overall drop a little bit in Q3. We think it's still going to be a grower long term. That's going to be 25%, 35% grower. We think it's really unique business. What you're always going to have are little bumps on the road where you're just not onboarding the same number of new clients during the given quarter or where an existing client pulls back a little bit during the quarter. And we saw that with a couple of clients. So we think it's, again, transitory. We think it's got a good top line. We think it's still a very unique business. It's going to grow into the future. So we don't think it's one where we're going to dial back the overall growth rate for Epsilon. We think it's going to be one of the drivers of Epsilon's growth rate in the future.

And just as it relates to Epsilon, obviously, I mean, how much of the Epsilon is tied to the secular growth – the positive secular growth trends within the marketing versus the agency? And as they see, the traditional agencies, obviously, are very much struggling when – and I kind of view Epsilon as being – having tailwinds from secular trends but what portions do and don’t. How do you do that.

25% agency like and 75% is getting the tailwinds.

All right. Thank you. Appreciated.

All right. Thank you, everyone. We'll talk to you next quarter. Bye-bye.

Thank you ladies and gentlemen. This does conclude today’s conference call you may now disconnect and have a wonderful day.