Bread Financial Holdings Inc (BFH) 2016 Q3 法說會逐字稿

Good morning and welcome to the Alliance Data third-quarter 2016 earnings conference call.

(Operator Instructions)

In order to read the Company's presentation on their website, please remember to turn off pop-up blockers on your computer. It is now my pleasure to introduce your host, Mr. Eddie 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 2016 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; Charles Horn, Chief Financial Officer of Alliance Data; and Bryan Kennedy, Chief Executive Officer of Epsilon/Conversant.

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 will provide useful information for investors. Reconciliations 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. Joining me today is Charles Horn, our always prolific CFO, and Bryan Kennedy, as mentioned, President of Epsilon/Conversant. Charles will go first and update you on the quarterly results, and then Bryan will discuss Epsilon/Conversant specifically. And then I will wrap up with a third-quarter look and a full-year outlook, as well as our initial brief discussion as we look into 2017. So, Charles, take it away.

Thanks, Ed. It was a terrific third quarter, with double-digit growth in all profitability measures. LoyaltyOne in Card Services came in better than expected for the third quarter, while Epsilon lagged a bit on the top-line growth expectations but executed the important turn in adjusted EBITDA, which was flat for the third quarter after being down in the first and second quarters. As a result of Q3's strong performance, we are raising annual guidance for the second time this year, increasing revenue from $7.15 billion to $7.2 billion, and core EPS from $16.85 to $16.90.

We did not pass through the full $0.32 beat to third-quarter core EPS guidance because a large portion is timing related, whereby some earnings were pulled forward out of Q4 to Q3. In particular, for AIR MILES, we expect redemption activity to moderate in Q4 after a run in Q3. Basically, collectors were getting ahead of the upcoming expiry date. And for Card Services, our principal loss rates came in 20 basis points better than guidance in Q3. We expect to give that back in Q4, which will prompt some reserve build. Lastly, I expect tax rates to pick up a bit in Q4.

Turning to the repurchase program, we've acquired approximately 3.3 million shares under our repurchase program year to date, and have approximately $300 million of our Board authorization left in 2016. You will notice from this slide that we are making a big change in our shareholder return philosophy, with our Board of Directors declaring a quarterly dividend, yielding 1% with a record date of November 3. Going forward, we expect to use the balance of share repurchases and dividends to return capital to investors. Ed will talk about this some later in the presentation.

Let's flip over to page 4 and talk about LoyaltyOne. As previously mentioned, LoyaltyOne exceeded expectations for the third quarter, as revenue and adjusted EBITDA increased 28% and 14%, respectively. BrandLoyalty came in as planned, with revenue up 5%. AIR MILES revenue greatly exceeded expectations, increasing 43% from the third quarter of 2015 due to elevated redemption activity.

As you can see from this slide, foreign-exchange rates had a nominal impact to the quarter. AIR MILES issued increased 6% in the third quarter, rebounding from a 3% decline in the second quarter, and is on track for about 3% growth for the full year. AIR MILES redeemed increased 74% compared to the third-quarter 2015, as we saw a spike in activity in front of the 12/31 expiration date. Although expiry was introduced almost five years ago, this was the first quarter of elevated redemption activity related to it. We expect that some of the activity is just a pull-forward from Q4 to Q3, and that activity will abate somewhat in Q4, a matter of timing.

Now, the answer at the question everyone will now be asking me is what does this mean for the accounting assumptions around the AIR MILES program. The answer is: it's too early to know. We need to see if the elevated activity continues through Q4, and we need to evaluate which collectors are redeeming, as the breakage estimate assumes certain collectors will redeem and certain collectors will not redeem. Only the second group really impacts breakage estimates.

Fast forward to the end of the year, if we determine that a breakage adjustment is needed, keep in mind that the accounting assumptions for the program are not evaluated in isolation. Rather, several other factors and accounting assumptions, such as cost a mile, [lap] a mile, residual, et cetera, will also need to be reevaluated. These potentially mitigate the impacts of lowering breakage rates.

Moving to BrandLoyalty, we recently signed supermarket chain Lowes Foods, whereby we will manage promotional campaign at 75 Lowes Foods stores in North Carolina, South Carolina, and Virginia. Overall, we are making solid progress in the US. Let's move to Epsilon and Bryan Kennedy.

All right, thanks, Charles. Epsilon's revenue increased to $543 million for the third quarter, up 2% compared to the third quarter of 2015. Importantly, adjusted EBITDA came in at $135 million, which is flat compared to prior year, after two straight quarters of decline. So to give you an update on our business, let me first discuss the EBITDA results.

Our most critical focus at Epsilon for the last several quarters has been to alter the long-term trajectory of the cost structure required to support our business. And a big part of that has been the ramp-up of the investment in our India operations to fuel a global staffing model. I am pleased that adjusted EBITDA, after a Q1 decline of 22%, followed by a Q2 decline of 9%, has leveled off flat in Q3, even while still carrying substantial duplicative costs. While it's taken, perhaps, a quarter or two longer than we originally anticipated, I think on the cost front, we can safely say that our strategy is working and we expect to see that trend continue to play out positively in Q4.

So let's turn briefly to the revenue picture and a breakdown of the offerings within the Epsilon segment. I will start with agency media and services, roughly 25% of the total business. These offerings produced $138 million of revenue for the quarter, a decline of 5% versus prior year.

At the beginning of the year, this business line was one of our concern areas, and our objective was quite simply to lower the drag that it was placing on overall growth rates. We've made steady progress with each sequential quarter, from a decline of 23% in Q1 to a decline of 15% in Q2, and in this quarter, as you can see, we've now narrowed that drag to a mid-single-digit decline.

Two key drivers fueled this: first, after a slow first half, Epsilon's agency offering has generated a number of new wins beginning to fuel back-half growth, such as the agency-of-record win we announced this quarter with Del Monte Foods. Second, we've been working to pivot Conversant's agency media offering away from a commoditized offering with a dependence on large holding company agencies. We've got a sound plan in place, and we are progressing each quarter.

Specifically, we're shifting towards an emphasis on more direct client relationships, on midsized agencies, and on a data-driven digital media offering fueled by our unique assets and scale. Essential part of this plan involves the consolidation of all of our digital media on to what we call our core platform, which is the engine that fuels Conversant's data-rich high-growth CRM business. And over the next two quarters, we will have substantially migrated the agency business to this core platform.

What that does for us is important. It increases the quality and granularity of the audiences we can offer to our clients, and it enhances our ability to deliver the metrics that clients are looking for. So, we are on a good track to continue the transition of this business.

All right, let's turn to the digital and technology platforms business, the other 75% of our revenue. We grew to $405 million of revenue for the quarter, up 5% over the third quarter of 2015. Performance here was lower than expected, driven by two key offerings that I'll outline, one on the negative side and one on the positive side.

First, the negative: our Epsilon technology platform business consisting specifically of loyalty and database solutions was soft for the quarter, dampening otherwise healthy growth for this category. This technology platform area in particular has been the focal point of the cost initiatives I described earlier, as we have worked to correct this historic issue of being able to produce nice top-line growth without demonstrating equal or better profitability growth.

Consequentially, at the same time as we've been pulling back to focus internally on reining in costs, we've seen some external market conditions spike up that include price pressure and increased competitive forces, and in turn, we've missed some top-line growth opportunities. Although it was critical to get to the cost side of this equation first, now that we're beginning to see it play out, we should be in a better position to return our focus on growth as we move through 2017.

On the positive side, the second driver for the quarter here was Conversant's CRM business, which continues to perform very well. This offering was up over 50% compared to prior year and really led the growth for this overall category again. We continue to see strong demand with both new client wins and ongoing cross-sell into Epsilon's client portfolio, and as of this quarter, we've closed 36 year-to-date wins and a book of business that already exceeds our 2015 book by over 25% in potential value.

In summary, if you look in aggregate, the year we've put together so far for Epsilon/Conversant hasn't been quite what we'd hoped for, but we have gotten some critical things done and are moving in the right direction. We addressed the cost structure within Epsilon, where the demand by clients for hot skills wasn't matching up with a market willing to let us pass on those costs. And while we might have lost some momentum on the growth front due to that, this was the right priority for our business. And we focused on the ongoing transformation at Conversant while driving successfully the growth of our CRM business.

We're happy with the progress there, and really that's Phase 1 that we're now wrapping up. Phase 2 will be our focus as we enter and move through 2017, completing the agency pivot at Conversant, expanding our CRM offering, putting our shoulder back into raising the growth profile of our database and loyalty business. We're well poised for that with our deep, unique assets and an excellent global team to service our clients. All right, back over to you, Charles.

Thanks, Bryan. Let's flip over to page 6 and talk about Card Services. Card Services had a better-than-expected quarter, with a revenue of 26% and adjusted EBITDA net of 14%, compared to the third quarter of 2015. Now what really drove the overperformance, and there's really several things that I'd point to, first, gross yields came in at 26.8%, 60 basis points better than original guidance and within 10 basis points of last year. Essentially, the cardholder-friendly changes we made mid-2015 have burned in, and the related compression has dissipated. Second, we continued to drive operating expense leveraging, with operating expenses as a percentage of average card receivables dropping 30 basis points compared to the third quarter of 2015. Third, principal loss rates were 20 basis points better than guidance of 4.7%, and trends in delinquency rates are moving consistent with our principal loss-rate forecast for 2016.

Looking forward, we believe the [B to] Q3 loss-rate guidance is due to timing from seasonality and that we'll give it back in Q4. Conversely, we did see credit sales slow during the quarter. While we increased tender share with our brands by approximately 150 basis points during the third quarter, softness at several of our core brands dropped credit sales growth in our core to 3%, down from 7% growth in the first half of 2016. We expect that softness to continue into the fourth quarter.

Counterbalancing, we have had a terrific year with new signings. The Children's Place, Ulta Beauty, and Williams-Sonoma, to name a few. Two of these brands have launched, and we are seeing early and very encouraging results as we welcome new card members to the programs. We have seen healthy applications in the first few weeks, thanks to our new frictionless mobile credit acquisition capability, which reduces the times to apply by 25% and reduces abandonment rates more than 5 times.

Turning back to our solid loss performance, we have a good view into the rest of the year and expect that we will track in line with full-year expectations. We also expect, as vintages mature, we will see delinquencies normalize around the middle of 2017, with losses normalizing thereafter. Ed will talk about this a bit later. With that, I will turn it over to Ed.

Great. All right, if everyone could be on slide 8, it says third-quarter and full-year outlook. It has a bunch of colored dots on it to make it easy to follow. This is the commentary I will give on third quarter, full year, then we'll go into a couple of other items, and then 2017. So, and from a consolidated level, third-quarter revenue of $1.9 billion, up 19%, and earnings per share of $4.74 up 20%. Obviously, this was the largest quarter we have ever had in our history, and needless to say, very, very pleased at how strong a quarter it was, especially in this environment of low growth for the economy.

Obviously, the -- compared to guidance, we exceeded guidance by a lot. And I did want to break out our thinking in terms of how much to pass through and not pass through. We decided to break it up into three pieces. Clearly, some of it was timing, right? Our loss rate in Q3 was lower than expected, but we expect the loss rate in Q4 to be -- to get that back. And as a result, if you want to call $0.15 of the $0.30 that timing issue between quarters for loss rates, that's just how it goes. So we don't expect to recapture that additional $0.15.

The second is, yes, we're flowing through some of the overperformance into higher guidance. We've raised guidance twice this year and that's what hopefully we'll continue to do, as is our practice. And then, finally, the third is a little bit of, hey, we need a little bit more flexibility in Q4, specifically as it relates to our AIR MILES business in Canada. As Charles talked about, we are now approaching the first time in the program's history where we are looking at actually expiring miles from, I think it was 1992 through 2011. So you've got 20 years about to expire.

Now, everyone's been informed of this five years ago and has been informed since then; however, we did not know exactly what the behavior was going to be as we're approaching the final end date here. And as a result, what we found is there is volumes that have picked up fairly dramatically in terms of people wanting to look at redeeming their miles. And as a result, what we found is that it did drive a big spike in redemption activity, which does drive a little bit of profit through to the bottom line as redemptions go up higher than anticipated, and that's what you saw in Q3.

However, what we want to do is we want to make sure that the consumer is having a good experience, and it has been challenging. We are putting, therefore, many millions of dollars into customer care. We have hired a ton of folks to work the phones on the redemption side up in Canada. We've got management taking phone calls on the floor, and as a result, we want to make sure that the millions that we're putting into this customer experience effort gives -- that's where we want to have the flexibility in terms of some of the excess from the Q3 overperformance.

And so, will we need it all? Don't know, but it's better to be prudent, it's better to have a good experience up there, because frankly we're getting flooded with folks trying to get this stuff in. And our job is to make sure that we don't damage the brand. So bear with us through the rest of the year. We think we're in very good shape, but we do want to have a little bit of flexibility here to make sure we can bulk up those resources as needed.

So, we are raising annual guidance to 11% growth and top line of $7.2 billion and core EPS of $16.90, or 12%. Again, I am reminded of the beginning of the year when we started out, and there were lots and lots of concern about the ability to continue to be a growth company in the face of normalizing credit conditions followed by, frankly, a very soft Q1 of only about 5% growth.

We knew the engine was going to pick up eventually; it did so starting in Q2. There were then concerns about whether that can continue forward. Hopefully, this has finally put that thing to bed, and what you have is you're beginning to see what this model can do. And what this model can do is double-digit top line, double-digit earnings growth, while also absorbing almost 10 percentage points of growth rate from normalizing and losses. Said differently, if we are guiding to 12% in a normal -- in the environment that wouldn't have been normalizing, we would have done over 20%. So, that's what the model can do. We're very happy with it.

And finally, on the share repurchase side, I think we've done around $700 million so far, and corporate leverage ratio remains quite manageable at less than 3 times. Again, Charles hates when I use -- when I round up or round down, but in general, you've got about $5 billion of [debt], you got a little under a couple billion of EBITDA. And again, the EBITDA is measured for our covenants and indentures across the entire enterprise, so you're talking well under 3 times, and that we feel is a very manageable level. So we are good there.

And then, for the first time since we started the Company, we are, in fact, establishing a quarterly dividend at 1%, and that is to provide a more balanced return to shareholders. So you're going to see both dividends and you're going to see share repurchases. We are a growth company, and that something that, frankly, I've debated for years and years and years. The tide certainly has swung in favor of let's do a combo platter of both dividend and share repurchase. That's what we're doing. And being a growth company, one of the nice things is you have a dividend that's going to grow quite nicely in the coming years as we continue to grow at a rapid clip. So it's sort of a nice thing to have out there and it's also not significant enough to our cash flow that it hampers our flexibility when it comes to either buybacks or M&A or additional growth in our card business.

Speaking of which, let's go to our card business, another heck of a quarter. Portfolio growth over 20% from both the quarter and the full year, and that's great news for me. What I am more pleased with, frankly, is the fact that our tender share, which, again, is defined as the amount of sales at a retailer that flow through our card versus other forms of payments. Basically, it means that more and more of the dollars that are being spent at retailers are flowing through our card. And that either means that we're bringing on more and more folks. Perhaps they're folks with -- of lesser credit quality; that's always been a question out there. And as we look at it, we continue to see that's not the case, and in fact, that 85% of the additional tender share or growth that's coming on our cards are coming from accounts that have been with us for three years or more. So it means that the personalized data-driven marketing stuff, so to speak, works.

We've had tender share gains of between 150 basis points and 180 basis points. What does that mean? It basically means if the retailer is doing a couple of points of sales growth, well, we're going to do somewhere between 5 to 8 points higher than that. That's how important this tender share growth is, and that's why it's so important, especially coming from those nice, mature accounts.

Gross yields, Charles talked about. Again, that was a huge concern of folks at the beginning of the year, when year over year we saw yields down over 200 basis points. We felt comfortable that, that was going to clean itself up as the year progressed. And, sure enough, by Q2, it was only down 80; in Q3 we're almost back to flat. So that's one big concern that we can check the box on.

Finally, delinquency rates and principal loss rates are tracking around the 50 basis points for the full year. We talked about Q3 loss rate was below our guidance of almost a year ago, and Q4 loss rate will make up the difference. But overall, you're going to slosh around and come in right around where we wanted to be for the year.

The other huge thing that got lost this year was the fact that this was, by far, the best year in our history in terms of signing high-quality clients, and it was by far our biggest vintage we've ever signed, vintage meaning that most of these folks are starting from scratch. We ramp them up over the process of three years and, when fully ramped up, they should be adding north of $2 billion of portfolio growth. And if you look at the clients Boscovs and Hot Topic, Forever 21, The Children's Place, Bed Bath & Beyond, Williams-Sonoma, Century 21, and Ulta, you're talking about, by far, the best vintage in terms of size and quality that we've ever had, which suggests the market is still quite robust.

All right, let's go on to the next page, talk a little bit about Epsilon/Conversant. As Bryan talked about, I would say if there was something that was a bit of a disappointment so far this year it would have to be our top line at Epsilon. Frankly, we were hoping that we could pull off the cost restructuring and the whole India initiative, while at the same time not losing any momentum on the revenue side. I think our appetite was a bit too large on that one.

So the revenue growth is still choppy; it was down 2% in Q1, it was up 5% in Q2, it was up 2% in Q3. That's not our model; that's not what we're shooting for. We want to get into a very solid mid-single-digit growth run rate on top line, and actually have a very consistent flow-through with some leverage to the bottom line. And in order to do that, as Bryan talked about, we needed to completely revamp our cost structure, essentially trying to mitigate the cost of the high skills with perhaps potentially some better leverage on the labor side. And that was the whole India initiative where we have our office today and are very happy with how that thing's ramping up.

So, from a longer-term perspective, from my perspective, from Bryan's perspective, we believe we have the cost structure fixed. And you can see that it's beginning to actually show results with your EBITDA, year over year, going from a minus 22% decline to minus 9% to flat. And I expect this thing up 5% in Q4, and then we exit the year and hopefully put this thing behind us. We will have the cost structure in shape so that we can now go after the top line, knowing what our costs are. So, it's been a long process and somewhat painful at times, but nonetheless, we're done with it.

We turn now to LoyaltyOne, which comprises our European-based BrandLoyalty business. What can you say there? They're continuing to track double-digit growth in revenue and adjusted EBITDA for the year. The thing seems to be a machine. The Canadian rollout has been very successful, and we're looking at $45 million or so in revs just from that, and that's only a year or so old. And then we had a big announcement of our first client in the United States, and that rollout is beginning to take off. And with that, hopefully we'll have many more announcements moving on from there. Obviously, with the US being 10 times the size of Canada, the opportunity there is quite large.

Looking at AIR MILES, which is our Canadian loyalty business, we are seeing double-digit growth in revenue, some of that driven by the extra redemption activity that's flowing through. But we are seeing mid-single-digit growth in EBITDA as well. So the program itself is vibrant, it's robust, it's still growing, it is far from sunsetting, that's for sure. And probably the only cautionary that we have there is we are working very hard to make sure that the brand does not get scarred with some of the drama around the expiration of 20 years' worth of miles. It's happening, it's going to happen; it's one and done for us financially. We don't think there's much risk there; however, we are being very cautious in terms of making sure we put as many resources as we can against this thing from people et cetera, et cetera, to make sure that the consumer experience is as good as possible, even though it is a little bit crazy right now.

So, that's where we are, and let's now talk about the 2017 outlook, again, at a very high level. As we usually do this time of year, we give a base case, and the base case is plus 10, plus 10. We expect also nicely growing cash flow, free cash flow to go along with that, which you'll tend to get with our type of business models.

That's the base; it doesn't assume any meaningful share repurchases, any meaningful M&A, and it also factors in what we believe is the final drag from the Card Services loss-rate normalization. So again, you're talking we expect to grow double-double, despite absorbing another 10 points of potential earnings growth that's going toward this normalization process. And right now, from what we're seeing, it looks pretty good out there. The consumer looks good and we expect another solid year of double-digit growth.

So we started 2016 with a lot of folks skeptical that we could, in fact, hit double-double, and in fact, we are doing quite a bit above that despite absorbing this drag. We expect a similar rollout in 2017, and hopefully, people now feel comfortable that this is the type of model that can absorb this type of normalization.

On LoyaltyOne, what are the big things we're trying to do? With BrandLoyalty, continue what they're doing. It's pretty straightforward, especially in North America, let's keep going. There's a lot of wood to chop out there, so that's what we're going to be doing.

In AIR MILES, we are -- we expect to have a solid 2017. That being said, I do believe that we're going to get dinged a little bit on our brand in terms of this whole expiry process. Even though it will be over by the end of the year and will be on a very simple quarterly basis and you won't have the type of noise or drama, the fact is, we're going to have to think hard about are we going to have to do something to make sure that our customers still feel loved and respected, and that they're getting value out of the program. So we're going to watch very closely what the behavior is, what the reaction is, and we're going to try to mitigate anything that comes out that's somewhat negative.

On Epsilon, Bryan talked about what we've done this year. It's critical next year that Conversant continue exceptional growth in its CRM business. This is the data-driven business where we ingest SKU-level information from various clients. We then match up that type of behavior to unique IDs associated with folks out there and reach out to them with very targeted messaging. That business has been growing rapidly -- what, 30%, 40%, 50%, something like that? Over 40%, and it's a big business, it's getting bigger and bigger. You're growing somewhere between $80 million and $100 million a year of revenue, and we expect that to be a big engine going forward.

What's lost in that is the whole agency piece of Conversant, which again is when it's the old value-click business where we used to provide inventory to the big holding companies on Madison Ave. That, of course has gone the way of the rotary dial phone. And what we're now doing is we have made the decision that this is absolutely a business that can be pivoted towards CRM Lite. Or for those clients who don't need the full, heavy-duty CRM, if it's just a list-activation-type program where we're reaching out very quickly for a sale that's coming up in two days, this is a perfect way of doing it. So we think we're going to -- we've got a solution there that could also help feed, eventually, into the big CRM business.

On the core, Epsilon core, we need to convert this new cost structure, quite frankly, into revenue growth. It's time to get going on this thing. We've got the cost structure that we want, we know how to price the deals now, and we expect to get some movement here on Card Services.

It's going to be a little bit different in 2017. The focus will be laser-like on existing organic growth and making sure this huge book that we signed in 2016 has a very strong startup. So we'll probably be a little bit less on just looking for bulk out there in the marketplace, that is, big portfolios to purchase. We're looking more on taking our bread and butter, which is these great names that have never had a program, starting them from scratch, nurturing them, and getting them moving and grooving. And we expect that allows us to keep the types of returns that is unique in the marketplace. So, that's what we're focused on.

Also, the less of a focus on pure bulk relates to the fact that some of these big files that are out there, frankly, the frothiness in the market is such that it's just not attractive to us, so we're going to be very disciplined about it, been here before, we'll do it again. Nonetheless, we expect very solid growth in cards, and we did not forget, talking about the final stage of normalization in credit quality, which should take place this year.

I did want to mention very quickly, I've seen a couple of commentaries out there about the 15 vintage and loosening credit and everything else. Let me make it very clear -- we did not loosen credit at all in any of our vintages. What happens, however, as you get further and further away from coming back from the great recession is that the clients that you sign and the cardholders that sign up will tend to skew more towards the lower end of our acceptable range. But we did not lower standards at all. But what you have is more folks coming from the lower end of that acceptable range, and it's not being balanced out at the higher end.

So as we moved into 2016, in the 2016 vintage, you'll see that we did, in fact, try to balance that out a little bit better. And so we did tweak the credit quality to be a little bit higher, and that should balance out nicely. So, I wanted to make sure that no one misunderstood that piece of it.

Okay. I appreciate everyone's patience. Let's go to the outlook with the fabulous chart that we have here. It says, delinquencies and net loss rates. Charles and I and the folks at Card Services have spent I don't know how much time trying to figure out the easiest way to communicate and to provide comfort that this is not a runaway train and that, in fact, this is a normal process that needs to take place after a long period of abnormally low losses following the great recession. And we have been through every single possible scenario and have come up with probably the easiest way for people to view this.

This is how we run our business, frankly, which is you look at delinquency rates. So an account that is delinquent needs to flow through 180 days of being delinquent before it actually writes off, and then the write-off results in a charge to the P&L statement. And that means that you have a nice view into what the ultimate loss rate will be.

Delinquency rates have been the best predictor that there is in the card business and in our business for sure, with a 90% correlation over many, many, many years. And so you will see from the first chart here that, in 2016, we saw our delinquency rates go from 4.2% to 4.7%, and you saw our corresponding move on the loss-rate side. We expect 2017 to be around a 5% and stabilize out there going into 2018 as well.

How did we come up with this stuff we just completed? We go through a portfolio by portfolio, over 155 clients, vintage by vintage in each portfolio, take a look at the aging, and then we lay out what we believe the delinquency flows will be. This is what we are going to use. We will update this every month. We will provide this in our monthly release, and this is really the best metric that we know that really captures everything. So if you start seeing these delinquency rates start spiking higher, then chances are you're going to see losses as well.

All right, let's go to our final page, I believe, or second to final page, which is the 2000 outlook (sic - see slide 12, "2017 Outlook") what we're calling closing the wedge. This is a very simple concept. What you are looking at here are delinquency curves, as fascinating as they are, let's spend a couple of seconds on it.

And what you'll see here is it's by month. What you'll also see is how the curves look very similar and how they go up and down, which means there is a great deal of seasonality to delinquencies. And so, for example, when we released our recent monthly data and it showed a 5% or something on delinquencies, and I got a couple of folks saying, oh my gosh, your delinquencies are going up -- that's seasonality. October, which we haven't released yet, you're going to be in the low 5%s, and then November, December, you're going to start dipping below into the 4%s; that's just the way the cycle works. So, again, this is the best way to look at things.

And if you look at the orange line, if you can see it, that's 2016. It shows how it is above the red line by about 50 basis points, which was 2015, and fairly consistently so. And that's right, the good predictor of your loss rate. As we going to 2017, this is the critical point, 2017 is the dotted blue line. You will see that we will enter 2017, we believe, around that 50- or 60-basis- point spread versus last year. And then, the whole bet comes down to, does this gap start to narrow? Does the wedge close?

If, in fact, the wedge closes over the next couple of quarters and starts to narrow, then you just know by definition that the eventual losses that are spit out based on those delinquencies will normalize as well. So what this essentially says is the normalization process is nearing its end. We get into Q1. We get through Q2, Q2 it should start to narrow, and by the end of Q3 you're basically on top of where we were this year, which means the losses that come out after that are going to be flat to this year.

And what does that all mean? It means earnings accelerates from the low-double digits to the 20 plus that you're used to. So, get used to the wedge; that's what were going to be talking about. And it also takes a lot of the noise out of the system and the idea of talking about delinquency flows up and down and master trust and growth losses and net losses and total losses and everything else, frankly, the noise level, it's hard to really get the message out there.

And so this is how we run our business. If the wedge doesn't close, then you know we have another year that we've got to fight through this thing in 2018. We do not believe that's the case, but we're trying to be as transparent as possible.

All right, finishing up on summary, over the past 15 years, this Company has had revenue growth of 16% a year for 15 years. Core EPS has been up 26% each year for 15 years. As we look at 2016 and 2017, because of the normalization and loss rates, which essentially means, look, the great recession washed a bunch of stuff out. You had pristine credit only coming out of the recession, and that's beginning to normalize now as people have repaired their credit and we have a more normal consumer mix in our portfolio. It's nothing to get worked up about; it just means things are returning to a level that we had anticipated. But the normalization process has knocked 10 points off of our earnings growth and brought it from the 20%s into the 12% we are looking at for 2016.

Nonetheless, despite this normalization effort, you have a model that's sufficiently diversified. And business is cycling at different times where we can absorb this hit and still grow double digit, as we have this year, top and bottom, and double digit in 2017, and then back to the 20% after the wedge closes.

At the end of the day, I will finish up finally, and say that I think the narrative for this Company, frankly, has been lost this year in terms of the noise that has been out in the marketplace. We expect to regain the narrative, and we will be talking more and more about the strategic positioning of this Company, why we believe financially, even in a normalization period, we're double-double. When the wedge closes, we go back to 20%.

But there's a reason for that. And the reason for that is all of our businesses are benefiting from the secular trends that we're seeing, whereby data and unique data is being used to gain insights to then provide marketing and creative that are then turned into personalized, one-to-one communications to our client's customers. And that's through the various digital channels where we have unique IDs on individuals around the world.

That's a secular trend we expect to continue. It's certainly not going away, and it's something that we're very excited about across all of our businesses. And as a result, our time -- as long as the wedge is closing and we're on track for our double-double, we will be spending more time trying to explain what really makes the model unique and has been so successful for 15 years. So that's it. I'm going to wrap it up and turn it over to Q&A. Operator?

Your first question comes from the line of Ashish Sabadra with Deutsche Bank.

Good morning. Thanks, solid beat. My question was around Epsilon. So Ed, you talked about it being choppy and you're taking all the right measure around the cost structure. My question was can you also talk about the demand environment and the competitive environment? Is there anything going on there which has caused the softness in the technology revenues, that moderated quite a bit from 15% to 5%?

Sure, I will take that question. I think, I mean that something that we look at pretty carefully. There's no question that there are a lot of companies that are playing in this space and have entered the space from the big Googles and Facebooks of the world that provide solutions to the SaaS players to the big agency holding companies that are scrambling to put together technology and data solutions.

Now when we stand back and look at it, we think that it's still a very vibrant market. We have a particular focus, which is to bring our data-driven solutions together with technology and creative and then deliver those in a client-intimate way. So we are looking for the segment of the market where clients really value a services-rich delivery. That's not going to be the model that the entire market will chase after, but we think that there's plenty for us.

So, I think for Epsilon at this point, this is a bit more of a scenario where our win rate has slowed a bit. That, perhaps, is because we've been more picky and cautious with all of the focus on costs. And then if you look at our existing client base, we've seen a bit of a pullback in some of our large clients, again, perhaps that's a loss of focus. That's something that we've had to work out as we move through 2017.

We certainly will have a much more intense focus on growth as we roll into the year. So, a couple of factors out there, but I don't think major drivers for this market.

Thanks for that color, Bryan, very helpful. Second question, Charles, was on the delinquencies. On the slide 11, you talk about -- or the slide talks about delinquencies going up 30 basis points in 2017. That's lower than the expectations for the charge off to increase by 50 basis points. I was just wondering if you can provide some more color there, and how much confidence do you have in the delinquencies only going up 30 basis points?

It's one of those I'd say we have pretty good confidence in what we have here, Ashish. As you can see, as Ed talked about, we built this up portfolio to portfolio. We do expect, as you look at the gap on page 12, it will be most pronounced in the first two quarters, then narrowing in the third and fourth quarter.

Going back to what Ed talked about, if you look in the trust, which is not a great illustration, but it's a little bit of a sample, you can see the 2015 vintage went quickly to loss; the 2016 vintage is going much slower. It takes about two years for an account to get what we call mature and less risky. So that 2015 will be flowing through by the middle of 2017, and that's why you'll see that wedge close quite appreciably.

So I'd say right now, that obviously we're looking forward 12 months, more than 12 months. We turn the portfolio every six months. So we can get pretty good visibility to it and we can make changes to the portfolio as necessary to track to the numbers we're looking for.

Yes, the way I look at it, it's just math, right? If you're going to be up, you're at 50 basis points, which is what we're talking about at the beginning of the year, and then the gap starts to narrow. And then you are running in the back half at almost flat to where you are, by definition, you're not going to be up your 50 or 60 basis points because of the math.

Okay, no that's very helpful. Thanks for that color. And maybe the last question on the -- if I -- based on the FY16 guidance if I back into the fourth-quarter number, that implies a softness, like the revenue moderates to 8% growth for the fourth quarter. Is that just conservatism or are there puts and takes there that we need to be aware of?

It primarily goes back to what we talked about with LoyaltyOne AIR MILES. We had a big ramp in revenue in Q3. We think some of that's a pull forward out of Q4. So what I would expect is the AIR MILES revenue to drop in Q4, and that's really what you're seeing sequentially.

Okay. That's helpful. Congratulations once again on the solid quarter. Thanks.

Thank you.

Your next question comes a lot of Wayne Johnson with Raymond James.

Hello, yes, good morning. Ed, I was hoping that we could touch base a little bit on the card services and the four growth components, namely same-store sales growth, tender share, new account wins, and then portfolio purchases. In the past, you've parsed that out when we've asked about it and I hoping you could do the same again as we look forward into 2017.

Sure be happy to, Wayne. I think one of the areas that is of concern is the fact that our client base, although the consumer, right, is doing quite well and is holding up the economy, what we're not seeing it we're not saying that translate into a lot of good growth at our clients. What we're seeing at our clients probably overall is actually their same-store sales, Wayne, have been down 1% to 2%, and that's not new news to anyone.

But soft good apparel, high-end home furnishing, jewelry, stuff like that you're just not seeing the consumer spending there. The consumer is spending in on autos and things like that, which, of course, benefits our auto business.

But the same-store sales are probably minus 1, minus 2, and as a result with the tender share gains, we can get probably plus 5, plus 6 in sales growth from the core. Whereas before, we used to do plus 10 when the clients were doing plus 3.

So, you've down-shifted a little bit in terms of expectations from the core. We still expect to get somewhere between 5 and 8 percentage points above same-store sales of the clients, but right now we're seeing softness. I know the NRF came out and suggested we're going to have a pretty decent holiday; hopefully, that is the case.

But again, that sort of down-shifted. Mitigating that for us has been the fact that the vintage that we did sign for 2016, frankly, was well north of what's going to be a $2 billion add to the file. The big growth chains like an Ulta and the Forever 21s of the world, those are going to do quite well. The other ones are very large clients, the Williams-Sonomas of the world, Bed Bath & Beyond of the world, they are going to be good size files. And so, I would say that the weakness on the core client base itself is going to be mitigated by the ramp-up on the organic side from the big 2016 vintage.

As we look at the portfolio purchases in 2017, frankly, we don't see much out there that is hugely attractive. If we do something, it will have to meet some very strict hurdle rates. We are not going to dilute the business. We want to see our yields hold up very firmly through 2017, so we may sacrifice a bit of growth on the portfolio to maintain more robust yields and a better use of capital.

So look for 2017 to reflect a little less robust growth from the core due to the core itself, the retailers being sub-par growth on their sales. But we do expect the tender share gains and very strong 2016 contribution.

I appreciate that response. That's very helpful. And just a quick follow-up, if I may. So, on the tender share gains, can you just talk a little bit about where -- what the tender share is for a mature retail customer, a customer you've had three or four years or so? What's that wallet share versus a new client when they're brought onboard and what's driving that cadence?

Sure. Fair question. Essentially, when -- for those of you not all that familiar with wallet share, it's essentially, right, the percentage of the retailer's sales that flow through our card. The first answer to where it is when we start the program, I can even do that without rounding, it's at 0. And then, over the course of three years, will probably get it up to around 30% or so of all sales at that retailer flow through our card.

And then the real fun stuff kicks in that the longer we have our client, the more SKU-level information we have on the customers of that client, which means the more precise the targeting is, when we do all of our targeting to the clients through all the different channels, whether it's point of sale or whether it's through all the various digital channels using Conversant.

And we actually have a number of clients who are over 50%. And what we're finding is a huge chunk of the incremental pickup for us, Wayne, is through the online channels. These retailers are using online very effectively, and our online sales are probably close to 40% of total card sales versus the retailer's, which are maybe 20%. You can expect to see the slow creep from 30% tender share up to as much as 50% over a period of 10 years.

Great. Thank you.

Your next question comes from the line of Andrew Jeffrey with SunTrust.

Hi guys. I appreciate you taking the question this morning. Just a question with regard to the wedge, Ed, which I think is a great way to frame up your expectations. When you talk about normalization, just causally, would you say that as the credit cycle progressed and you saw the 2015 vintage evolve that underwriting tightens a little bit, and that's why you're so sure that as those vintages season that the loss rates will moderate? Or is there something else at work? I'm just trying to get a little granularity on your confidence level there and what's driving it.

Yes, I will take half, Charles can take half. The bigger thing is that the losses -- in the great recession, you flushed out more than you normally would in a more typical recession. And as a result, what was left over, who was left standing meant that you had a lot of like super prime coming into the file, which is not really our bread and butter.

We are much more in the mid to upper prime type range, and what's happened over the course of the number of years the recession is in the rear-view mirror, you're having more and more folks who have repaired their credit, who are back in the market, who have jobs again, who are working, and they tend to flow into the file. And as a result, what you saw was it's less about losses going up; it's more about losses are beginning to conform more to the type of consumer that we cater to.

So that's the big misperception out there of we're not shooting for 4% loss rate. That's not how we optimize a file. When you're talking the folks that we want and the people who visit the stores, our loss rate, we believe, is optimized at a higher level, and those are the folks that have been flowing in. Specific to your question, yes, when we saw the 2015 vintage and we saw that we were kind of -- it looked like a lot of the prime folks who are the high end of the pool, there were fewer of those and more folks who still met our credit criteria but was a little bit lower in the overall pecking order, they tended to dominate a little bit more and we needed to tighten up a little bit.

That's exactly right. So not a lot I can add to it; it's just purely mix. A higher percentage of your applications in the early part of coming out of the recession are high quality individuals, super prime, because you've basically burned a lot of people out of the market.

As they start to repair their credit four or five years thereafter, they start applying for credit. They're hitting the bottom end of your underwriting criteria, so have a mix shift that you then adjust. If it gets beyond a certain spectrum, and that's what we did. 2015 moved a little bit quicker than what we thought it would, so we tightened up a little bit in 2016 to keep it tracking to the numbers we want.

Okay. Thanks, and just as a quick follow up, can you just talk a little bit about how you view Amazon with regard to its impact on your client same-store sales, recognizing the tender share mitigates some of the impact?

Yes, and whether it's Amazon or the whole online Amazon effect, what it's essentially done is two things. First, it has made shoppers much more educated when it comes to where they can get the best price. So whereas before, you had folks it was a little bit hit or miss. Now, everyone's got their phone, everyone's comparing prices. And as a result, that's what's putting a lot of pressure on our clients, which is really on the pricing side. And so that's why the margins have been so tough at our clients.

Obviously, the whole online space makes it a lot more convenient for people to work. I do think the retailers in our space are coming back strongly with their online offerings, and we'll make a nice dent on that side of it.

But Andrew, I really, from what we're seeing, it's the pricing side of it. It is -- your -- Amazon has educated the consumer to such a degree that the consumer, no one pays full price anymore. And as a result, your sales are suffering as much from pricing declines as from just unit declines. So that's what the retailers are struggling with is what is the new model.

What we're trying to do is, obviously, this flows very nicely into how we run our business, which is you can't just have a $200-million marketing budget and go out and splash it around to everyone. You need to be much more precise, much more effective with how you market to folks down to the one-on-one level. So this impact has been driving quite a bit of thinking on the part of retailers to change what has been, for decades, their way of going about marketing and moving towards the unique data-driven personalized approach we do.

Thanks a lot. I appreciate it.

Brett Huff of Stephens.

Good morning, guys, thanks for taking my question. First question on Epsilon, you talked a little bit about some of the things you're going through. And I think you said you needed to refocus on some of those big clients, that they've pulled back their spending and that's maybe there was -- there's a loss of focus as you're reducing cost. Can you talk about how that might compare to any structural changes that you see? Because you mentioned it wasn't structural, but I just, I want to make sure I understand that point that you made.

Yes sure Brett, I think from the perspective of the existing client relationships, obviously, you've got to do an excellent job of serving clients, walking the halls, and bringing solutions to them that really meet the needs that they have. And we strongly believe that demand for full-service strategic support is still rich in the marketplace. There are certainly alternatives with software as a solution offerings that would allow a client to maybe do something in-house.

Over the long haul, our bet is that the kind of skills required to really make that work and to drive results are difficult to create within a client's organization. So I think from a market perspective, having watched this for 20 years, I don't see that's going to change radically. So that's the focus point for our business.

In terms of whether there's something structural that goes well beyond that, we'll watch the quarter to quarter. But we really believe strongly and this is coming from the voice of the clients that we serve and the global 1000 CMO that we're targeting, that need is there in the marketplace and that those are skills they don't have in-house.

The focus issue for us and really the reason for putting this foundation in place is the change we did see is a scenario where the skills that you need to service those relationships aren't always matched up with the client's willingness to continue to pay more for those as cost of living goes up, et cetera. So that's really the driver behind our shift on the cost structure.

Great, that's helpful, thank you, and just second follow-up question on card. Ed, I think you mentioned more folks in organic growth, less buying portfolios. I think you've also in the past said that there are a couple of programs you're going to non-renew, and I wonder what the status of those were in your commentary of focusing on organic growth and if there's any knock-on effects from about non-renewal. Should we expect more of that over time from some of those clients or does that moderate?

So in the first quarter, we did put the two portfolios to held for sale, knowing that we were not going to renew them at the end of 2016. We would expect those two programs to be gone from ADS by the end of 2016. If you look at the average card receivables we give you, Brett, we excluded already. So we're not considering that in the growth rate. So I don't think it's going to have any meaningful effect going into 2017, especially, to Ed's point, given the very strong 2016 vintage we've signed to replace it.

That's what I needed. Thanks, guys.

Our final question will come from the line of Robert Mattson with Dougherty & Company.

Great, thanks for taking my call and congratulations on the quarter. Ed, quick question, the dividend, in just [a long time], the preference was to not do it for tax-efficiency reasons. A little thought here, is this, you traditionally broke up your cash flow into three buckets: M&A, funding the receivables, and then share returns. Will this strictly come out of the share returns or will this be incremental? In other words, should we see this as us maybe pulling a little bit out of M&A, a little out of funding?

And then also, you mentioned about it growing. Should we also be thinking of this as somewhat of a payout ratio? In other words, it should grow with the -- more or less with earnings?

So on the first one, Robert, I would say it just adds a fourth option that we can flex based upon what we see in the market. It could flex against M&A, it could flex against buyback. So it's really just now a fourth item we can consider in our capital allocation strategy.

Okay. And going on to Epsilon on a little bit different angle, less about the current customers. If you look out in the space, there's a lot of vendors, particularly on the technology side, for instance, Adobe, which is one of your partners. They're seeing opportunities to gain share. They're seeing a lot of activity. Our research shows a lot of activity on the digital transformations and the data-driven marketing. And they're seeing opportunity to spend to gain share.

Could you elaborate a little bit more, Bryan, the kind of activity you can do to bring the growth rates higher and make them more durable to get around -- away from this choppiness. It would seem to me that the growth rate should be able to get higher given the end-market strength and desires for (inaudible) down this path.

Yes, no question that right now, given all of the options in the space, it's a complicated decision process for our clients to navigate in terms of thinking through all of the -- those options and then selecting the right choice. Part of what we've done historically is position ourselves as a trusted advisor that has the ability to really help a client navigate those decisions, and then actually drive outcomes from their investment. That's a different position than a lot of the other providers, including some of our partners who are a little more focused on a software sale with less accountability and responsibility to really drive results.

So part of what we need to do better I think is double down on that bet in terms of educating our clients, and then focusing carefully on the players that really value the services that we bring to the table and not, frankly, chasing everything.

The other side of it, from our perspective, is when you look at the assets that Conversant brings to the table from digital execution perspective, from the ability to really have a pristine cross device capabilities and then incredible ability to reach consumers. When you put that together with our big technology platforms, including Harmony and our database and loyalty platforms, you start to really see, I think, some significant differentiation in terms of what we do versus the rest of the marketplace.

We have, frankly, not put our shoulder significantly into that effort. It's part of what we're working on right now. In fact, we have several key pilots with clients that begin to put those capabilities together in a more integrated way. And I think that's going to be a key effort for us as we move through 2017 that will start to drive part of what you're asking about.

Okay, and as a final question, going back to the delinquency trajectories, if we were to look at the ability to meet that wedge, missing the wedge, would that be -- is anything intrinsic to your portfolio or would this be largely a broader macro event?

This is the portfolio by portfolio roll-up, so again, we go through all 155 portfolios, every vintage in it, and we can see where they are in a life of the account and how the curves are looking and then we project out. So it does not assume any huge improvement on the macro side at all.

Okay. Great. Thanks a lot.

Okay. All right, so I think that's it. Thank you for all your time and we'll talk to you next time. Bye.

Thank you, that concludes today's conference call. You may now disconnect.