Bread Financial Holdings Inc (BFH) 2016 Q1 法說會逐字稿

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

(Operator Instructions)

It is now my pleasure to introduce your host, Mr. Steve Calk of FTI Consulting. Sir, the floor is yours.

Thank you, operator. By now, you should have received a copy of the Company's first-quarter 2016 earnings release. If you have not, 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 Pearson, President and Chief Executive Officer of LoyaltyOne. 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 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 would like to turn the call over to Ed Heffernan. Ed?

Thanks, Steve. With me today, as mentioned, is Charles Horn, our CFO, and Bryan Pearson, Head of our LoyaltyOne business, and that includes both the AIR MILES business in Canada, as well as the BrandLoyalty business headquartered out of the Netherlands. Charles will update you on the quarterly results, then Bryan will give you some color into the various businesses at LoyaltyOne, and I will wrap it up with the scorecard for Q1 results, as well as my outlook for the full-year 2016. Based on past comments, we are going to move with pace so we can have more time for questions at the end.

Take it away, Charles.

Thanks, Ed. Turning to slide 2, we will talk about the first-quarter results, and they were pretty much as we expected, with revenue up 5%, to $1.68 billion, and core EPS up 5%, to $3.84. It was a quarter with several positives but also a few negatives that tempered overall results. One of the negatives, foreign-exchange rates, reduced both revenue and core EPS by about 2 points for the first quarter. The good news is that FX rates seem to have stabilized, and the drag could drop to near zero in the back half of 2016.

Looking forward, we expect growth in each quarter to accelerate as 2016 progresses. Revenue growth of 5% in Q1 is expected to increase to 8% in Q2, and then double-digit growth in Q3 and Q4. Adjusted EBITDA and core EPS are expected to follow a similar trend. Now, why do we believe this? Putting aside improving foreign-exchange comparables, there were several reasons. First, gross yield compression at Card Services should lessen each quarter as we anniversary several program changes made last year. Second, loss rates are expected to drift downward from first-quarter highs. Third, BrandLoyalty has passed its most difficult comparable. If you remember, in the first quarter of 2015, the revenue was up over 100% on a constant-currency basis. Lastly, Epsilon's agency weakness should lessen, coupled with a record CRM backlog for second-half delivery.

During the quarter, we acquired 2 million shares under our repurchase program for approximately $414 million. The buy-back program remains our top capital allocation priority, and we expect it will be executed somewhat ratably over the remainder of 2016. Essentially, we will look to deploy our free cash flow while maintaining the leverage ratio of less than 3 times.

With that, let's go to the next slide, and Bryan will walk you through the LoyaltyOne results.

Thanks, Charles. LoyaltyOne had a better-than-expected quarter, as revenue decreased 2%, to $381 million, but adjusted EBITDA increased 10%, to $85 million, on a constant-currency basis. Breaking it down further, at AIR MILES, our revenue increased 4% and adjusted EBITDA increased a robust 11% on a constant-currency basis, and that was driven by a favorable issuance mix and prudent cost control. As expected, at BrandLoyalty, revenue was down 9% on a constant-currency basis, due to program timing differences.

Encouragingly, adjusted EBITDA increased 9% on a constant-currency basis, due to higher-margin programs that were driving a 300-basis-points expansion in adjusted EBITDA margins. AIR MILES issued increased a robust 5%, or 57 million miles, over the same quarter from last year. Issuance was particularly strong with our financial vertical, as we continue to convert [Vimo] MasterCard holders to the premium World Elite card that has a more compelling consumer proposition and a higher AIR MILES earn offer.

This strong growth was achieved despite economic uncertainty in Western Canada that is the result of the declining global oil prices that has been negatively impacting activity at our grocery partners, and also the suspension of point issuance on prescriptions in British Columbia. Looking ahead, we anticipate a 2% decline in issuance growth for the second quarter, due to reduced promotional activity and the one-time conversion of the [Sobe's] Legacy program that happened in 2015, which of course would not be repeated. As we've stated previously, quarterly results can vary based upon the timing of promotional activity, but our target remains a full-year issuance growth of mid-single digits.

AIR MILES redeemed increased 6% compared to the first quarter of 2015, with our AIR MILES cash program -- that instant reward option -- driving approximately half of the increase, as its liquidity feature resonates with collectors in the current macro environment. For the full year, we anticipate redemption growth to be in line with issuance growth. For the year, we anticipate full-year constant-currency revenue and adjusted EBITDA growth for AIR MILES to be in the mid-single-digit range, as well.

Now, turning to BrandLoyalty, we anticipate a very strong second quarter, whereby revenue could be up as much is 50% on a constant-currency basis. On the North American expansion front, BrandLoyalty continues to increase its presence in Canada, with six promotions that have been planned for 2016. And this represents a significant increase in sales over last year. In addition, we launched our first pilot, in the US, late in the first quarter. We are pleased with the results to date and look forward to additional programs in this market. The full-year expectation for BrandLoyalty continues to be mid-teens growth in both revenue and adjusted EBITDA, as compared to 2015, on a constant-currency basis.

Lastly, let me update you on dotz and Precima. Starting with dotz, our Brazilian coalition program, where we continue to see steady growth in our collector base to over 18 million currently. In November of last year, dotz successfully launched the coalition program in Rio de Janeiro, with the Cencosud grocery banner called Prezunic as a key partner. Cencosud is the third-largest grocer in Brazil, and collector reception has been positive, with over 1 million collectors now enrolled and active in Rio. Precima, our data analytics unit, continues to fire on all cylinders, recently signing a deal with Morrison's -- one of the UK's top-four grocers -- where we will apply data intelligence and shopper insights to help Morrison's more effectively apply their customer-centric retailing approach.

In closing, LoyaltyOne achieved solid first-quarter results, with a positive outlook for the remainder of 2016. I'd like to take this opportunity to congratulate the entire team on a great quarter, and look forward to their commitment to our valued clients and to LoyaltyOne. Awesome job, everyone.

I guess it's back to you, Charles.

Thanks, Bryan. Let's turn to slide 4. Epsilon's revenue decreased 2%, to $493 million, and adjusted EBITDA decreased 22%, to $81 million, for the first quarter of 2016. Results in this segment were mixed, as three-fourths of the business -- digital and technology platforms -- grew a robust 8% during the quarter, while the remaining one-fourth -- agency -- underperformed, where the revenue declined by 23% versus the first quarter of 2015. Adjusted EBITDA increased $23 million from the first quarter of 2015, primarily due to higher payroll costs, stacking levels increased in anticipation of revenue that has yet to occur.

Before we dive into the individual portions, you see that we have a new reporting format here. We felt that the old formats where we used to talk about database and data, and so forth, really is not reflective of the way we go to business these days. Specifically, as the broad portfolio of offline/online services Epsilon delivers are packaged or bundled omni-channel solutions, it just doesn't make sense for us to give it to you in that format. So what we've done is, we've put it together where we're going to highlight two categories that represent the underlying services and assets deployed to serve our clients.

The first category, digital and technology platforms, consists of businesses which is primarily derived from revenues tied to one of the many core data and/or technology platforms that Epsilon and Conversant manage. These platforms are used to house and manage data in real-time, to analyze and model data, and to automate the usage of that data across a variety of channels, whether it be through a loyalty program, within an e-mail communication, or an online message delivered on mobile or desktop devices. The remainder of the business, agency media and services, consists of services which are primarily focused on supporting the agency sector. Both media services deliver to agencies on behalf of their clients, as well as agency engagements in which Epsilon acts as the primary agency directly on behalf of its clients.

Now let's turn to the digital and technology portion of this segment, where quarterly revenue grew by $28 million year over year, or about 8%, mostly due to continued success of the cross-sell efforts and on-boarding of the strong CRM backlog within Conversant. The products offered within the CRM business continue to resonate with both new and existing clients, as 15 new contracts have already been signed this year versus 30 all of last year. The 2016 vintage thus far includes GNC, Brooks Brothers, Coach, and Sears, and is a great follow-up to last year's efforts. These clients, along with others, such as Walgreens, are connecting with real people as a result of combining anonymized first-party with third-party data and 80 billion daily online interactions, to develop a persistent consumer ID.

Moving on to agency media and services, revenue declined 23% from the first quarter of 2015. For Epsilon's traditional agency business, we saw broad-based weakness, especially in the telco, CPG, and retail verticals. In addition, the agency business at Conversant, which is the non-data-driven traditional display business, was down as large agencies push more programmatic buying through their internal trading platforms versus using third-party DSPs.

We don't know if this bubble is a bubble within agency or more structural, and needs to play out as the year progresses. If it's structural, we will have to be more aggressive on going after cost cuts to maintain our earnings. Although we will likely continue to see volatility within agency on a quarterly basis, it is a key part of the CMO selling process, which attracts more broad and less IT-centric relationships.

Now let's go over to the next slide and talk about Card Services. Card Services delivered strong financial results, achieving its 17th consecutive quarter of double-digit revenue growth. Revenue increased 17% for the first quarter of 2016, driven by a 27% increase in average receivables. Adjusted EBITDA net grew 9%, dampened slightly by the provision build. Importantly, principal loss rates were 5.2% for the quarter, 10 basis points better than our original guidance.

Focusing on the key drivers for our growth, credit sales increased a healthy 25%, driven by our proven ability to build customer loyalty for both our longstanding and newest brand partners, as well as favorable weather and an early Easter. Our core business, brand partners with us for four years or more, continues to contribute roughly half of our credit sales growth. Importantly, we continue to grow tender share, up over 200 basis points for the first quarter, by leveraging our large understanding of our brands, along with the deep data analytics we bring to bear.

Most of the tender share pickup is being driven by card members who have been engaged with their card for multiple years, not from new accounts. In other words, these are tenured card members with a proven spending and payment history, and they're demonstrating increased spend with the brands they love. This is a testament to the power of our customized marketing programs and the value they deliver to our longstanding partners. It is also a key differentiator and critical element in our ability to welcome new partners.

Looking at the remaining drivers of our performance, we talked about revenue grew 17%, driven by the strong AR growth. Importantly, we realized strong operating expense leverage, as we are able to drop our expense ratio by 70 bps for the quarter. As expected, gross yields experienced some compression, driven primarily by brand mix, accounting for acquired files, and lower late fees, as we have seen several declines for several months in our first-stage delinquency bucket. This pressure is predominantly in the first half of the year, and we expect yields to normalize as we anniversary these items.

Managing losses was key in the first quarter, and we exceeded guidance previously outlined. Losses, delinquency levels, and funding costs were all in line with both expectations and seasonal trends. During the quarter, we decided not to renew two card programs (technical difficulty) renewal at year end. The related receivables, net of the allowance for loan loss, have been reclassified to assets held for sale at March 31, 2016. Accordingly, $430 million has been removed from card receivables, and $15 million has been removed from the allowance for loan loss.

Our reserve for loan losses was 5.7% of reservable receivables at March 31, 2016, up from 5.5% at December 31, 2015. Reserve dollars dropped by $15 million from December 31, due to the reclassification previously discussed. In closing for Card Services, we delivered another stellar quarter, with all the key metrics outpacing the industry and consistent with our expectations. We are positioned for another record year, in line with our strategy of 20% plus credit sales and AR growth, leveraging both organic growth and new partnerships.

With that, I'll turn it over to Ed.

Great. Thanks, guys. Why don't we turn to first-quarter scorecard, and the goal, of course, is to have more green than yellows and reds. And right now, we are tracking with [the majority] being green. I would say, in terms of Q1, a couple of general comments. One is, as Charles mentioned, this was, obviously, our toughest quarter, and comp versus last year was very difficult. And, putting in a plus five top, plus five earnings was right on the money of what we wanted to do. And our view into the rest of the year looks solid. So it is good to get that behind us.

Also, you will find that it was a mixed bag, in terms of what had got us from A to B. But, at the end, the pluses outweighed the minuses and we got where we wanted to be. As we move into the second quarter and the rest of the year, things are accelerating nicely. We have good line of sight into the rest of the year. From a first-quarter perspective, consolidated, we came right in where we said, in terms of guidance, at about $1.7 billion top, at about $3.84 on earnings. That was -- that included getting hit on the FX side with about $26 million in top, at about $0.05 on earnings from FX, which, if it holds, will fortunately start to anniversary as we move into the back half.

We talked about the toughest comp of year is now behind us, and we were pretty active in the buyback, taking in a bunch of shares which we thought were pretty cheap. So, overall, consolidated, we got where we wanted to be, and remain on track for the year. And Card Services -- just the highlights, of course -- the portfolio continues to grow very quickly, with over 20% growth, revenue in the high teens. And then, obviously, due to the fact that we are growing so quickly, we need to obviously set up reserves, and that tends to be a damper on EBITDA. Nonetheless, that did grow 9%.

Yield gets the yellow. It was little over 200 basis points, due to multiple factors that we can talk about. Again, the good news there is that it is already beginning to burn away, and we expect less compression as each quarter goes by for the year. And we have pretty good visibility into that, as well -- so, more of a Q1 issue. The loss rate, we spent the first three months of the year here dealing with two questions that people were spooked about, the first being credit quality and the second being Canada going into a depression.

And what's interesting is, those are the two items that probably performed best during the quarter. Credit quality actually came in a little bit better than we had anticipated. We now expect it to drift downwards, as the year progresses, from a seasonally high Q1. And, similar to what we said three months ago, we will say the same thing now, we check with all our collectors and everyone else. There is zero pressure that we are seeing coming in that would suggest that there's any type of stress on the consumer at this point. So, we're just not seeing it, we did not see it before. And again, we are just normalizing the loss rates in the file, which right now we are tracking nicely to that.

I will emphasize the next point, which is tender share. That's often lost in all these discussions about growth, and are you growing too fast, too slowly? How are you growing? Is it good quality, bad quality, and all that other stuff. Tender share is up 200 basis points, which means if 30% of the sales at a retailer were on our card a year ago, now it is 32%. And that is not being moved by putting on a bunch of new accounts. And, in fact, what it essentially means is that when the retailer is growing 2%, 3%, we are growing 10%.

And so the first 10 points of portfolio growth are coming from a deeper penetration of that retailer sales base. And the next bullet point, to me, really brings it home, that 80% of this tender share gain, they are coming from the mature accounts. They're coming from accounts that have been with us for 20, 24 months or more, which means they've passed the peak loss point and they are very mature, and our ability to get in there and motivate incremental spend is what it is all about. And that's the data-driven targeted marketing.

And I think that's probably the most important takeaway, is that this is not coming from tossing on a whole bunch of new accounts or changing underwriting or anything like that. It is coming from driving those very long-term loyal customers to visit more often and spend a bit more on each visit. So, that continues to work out well. The pipeline, we expect another year of $2 billion in new vintages. We have a bunch that have been signed, have not been released, and they will be released over the next couple of quarters. So, the pipeline looks good.

And, as we move to Epsilon on the next page, that's where we run into -- probably, if you were to ask me about where the concern is for the year, it is certainly not on the credit side, it is certainly not on the Canadian side, it is certainly not on the technology and data and digital sides of Epsilon and Conversant. We are focused, however, on some weakness on the agency side, which is about one-quarter of Epsilon and Conversant. And what we saw there was a general pullback in a number of the verticals, and we're going to watch that carefully, as Charles said.

We don't know whether that's just a result of CMOs taking a breather in the first quarter, given the turmoil, or whether that's more structural. In any event, if it is structural, we will make sure that earnings will remain on track, which means we'd have to focus on the expense side. But we don't know right now, and we are not going to knee-jerk it at this point. On the other side of it, the digital and technology platform stuff, up 8%. That's three-quarters of Epsilon and Conversant, and we continue to see very, very strong take-up, in terms of the data-driven targeted display offerings from Conversant.

The number of wins that we are seeing right now are already half of what they were for the entire year last year. And then, from an expense perspective and flexibility perspective, our India office is up and running, and we are hopefully continuing to staff that up as the year progresses. And that will finally begin to move the needle on the expense side towards the latter part of the year.

LoyaltyOne, I think Bryan covered that. Obviously, BrandLoyalty is a very choppy business. It had, frankly, a little bit better Q1, from the EBITDA perspective, than we had initially talked about. But recall last year, Q1 was when revenue was up 100%. So it was then down in Q2 this year, it will be more of the reverse, where you will have a very strong Q2 for BrandLoyalty, and it is all driven by when the grocers have decided to put the programs on the calendar. Bryan also talked about the BrandLoyalty expansion into North America. Canada is doing quite well, and moving nicely towards our goal there. And the US, actually, we are beginning to make some nice inroads there, as well. So good news on that front.

Again, the concern on the AIR MILES side in Canada was that issuance would fall off as consumer spend was weak. What we found was the reverse, which was issuance was very strong, and specifically issuance in the financial services space, which had been lagging before, actually came back quite strongly. So it looks like Canada is still alive and kicking, and doing quite well. Revenue and EBITDA, plus 4%, plus 11%, frankly, on a constant-currency basis, was the best we've seen in years. And so it looks like the business up there is back in growth mode.

Overall, for those of you who have covered ADS for a number of years, it is a typical ADS quarter. You have got a bunch of things that went a little bit better than expected, you had some things that went worse than expected, and the end result is they even out, and we meet or we beat our guidance. And that's pretty much been our track record for the last 15 years. We expect that to be the case going forward, as well. If you look at it in the Cards group, the yields were little bit worse than we had anticipated, but losses were little bit better. And then at Epsilon, the digital and tech platforms remained on track.

We talked about agency and media was weak. That was offset by the fact, in LoyaltyOne, that AIR MILES was better, while BrandLoyalty was tracking. So, again, with us, you can pick and choose whatever data points you want, but at the end of the day, the pluses and minuses hopefully even out to get to a double-digit growth rate for the year. And as we turn to our full-year guidance, that's exactly what we are saying. It is very straightforward.

Now, with Q1 behind us, we have even a higher level of confidence in how the rest of the year is going to play out. And we feel very comfortable reiterating our full-year guidance of revenue of $7.1 billion, up 10%, and earnings of $16.75, up 11%. And the nice thing about it is the slowest quarter is behind this. We're going to ramp up to high-single digits in the second quarter, and then we will do double-digit top and bottom Q3, double-digit top and bottom in Q4, and that will give us a nice jump-off into 2017.

Visibility is quite strong. Q2 guidance revs around $1.6 billion, up 8%, and earnings also up about 8%. I think those are good numbers to start the acceleration. From a Card Services perspective, you will see the principal loss rate forecast that we gave a while ago. And, lo and behold, it seems to be playing out at that level, if not just a smidge better, which is good. We expect, for the year, Card Services to once again deliver double-digit growth in revenue and adjusted EBITDA, net of funding.

Yield compression, which I'm sure we will get some questions on, will lessen each quarter as the year progresses, so that we wind up at the end of the year with a very modest compression, which is how we want to take it into 2017. And then we said the pipeline is very robust, and that will get into a little bit about our decisions around who to renew, when to renew, at what levels will we renew clients, given that there's a certain amount of capital we want to put into the business, but we want to maintain a 20% plus growth rate in the file. And that means maybe our capital is better spent on other areas, and we don't renew certain folks.

So that's where we are on Epsilon. We talked about digital and tech platforms are three-quarters of the business. We expect that to continue clipping along in the high-single digits. The signings for the Conversant CRM vintage, if you recall, last year, was about 30 wins and a very strong vintage signing of 80-ish plus, once they fully spooled up. We expect at least that this year, if not a little bit better, which would be great.

And then the agency stuff, look, we are going to watch it, and we are going to see it is a people intensive business. It is at about one-quarter of Epsilon and Conversant. From what we are seeing, it looks soft, but it looks already like it is lessening, in terms of the softness, as we progress into Q2 and the rest of the year.

Okay, last slide, full year, LoyaltyOne, BrandLoyalty, Q2 is the big one for them. So expect a very, very strong double-digit rev in EBITDA. And then, in AIR MILES, we remain comfortable that issuance, which is the key metric there -- it is how we get paid -- is going to be up about 4% for the year, which will drive, on a constant-currency basis, low-single-digit revs and EBITDA full year, which again would fly in the face of the weakening consumer in Canada. Overall, as I mentioned, several moving pieces, but full year is shaping up nicely. We're seeing zero stress on the consumer side. We are not seeing weakness in Canada. So those two data points are consistent with last quarter and the quarter before that.

I just would talk a little bit about the model itself for Alliance. From 2007 through 2015, we produced annual growth rates of 15% and 18% for revenue and earnings per share, respectively, every single year. And for 2016 and 2017, as losses normalize, and with the huge growth in the file, we're setting aside reserves for future losses. What you're essentially doing is, you're knocking that growth rate down a bit, but you are still looking at low end double-digit growth rates in top and bottom. So this year, we are looking at 10% and 11%, and that includes [some drag] from FX.

And then, in 2018, if in fact we are right and these rates have stabilized, then there's no reason to doubt that it would snap right back to the prior model, and you'd have a re-acceleration in growth rates. So, if you look at long term for the Company, you are looking somewhere mid-teens to high teens, revs and core EPS. Then, when you are in a period of normalization and building reserves, you are probably looking at low end, top and bottom, double digit, and then you snap back mid-teens when that normalizes.

So, overall, not too shabby of a model. You will also find with us that we tend to have different levers and different businesses cycle up and down at different times. But, overall, the model holds together quite nicely. For guidance, on the last slide, we're keeping it the same. And we are not going to spend a lot of time going through the [nits and nats] of how we get there, but it looks pretty solid at plus 10%, plus 11%. And that puts our organic growth rate actually quite a bit above our 3 times GDP target.

We looked at cash flow, looks very nice, very strong this year, at about $1.4 billion. And, look, when it comes to FX and all that other stuff, we will adjust it when it becomes meaningful as the year plays out. But, again, we are not going to be tweaking everything every quarter. We are going to try to get this thing going on a consolidated basis.

So, that's all I had. Why don't we open it up for Q&A? Operator?

(Operator Instructions)

Sanjay Sakhrani, KBW.

My first question is on the agency side of Epsilon. Charles, you mentioned that you are not sure if it is structural or not, in terms of the weakness there. And maybe Ed or Charles, you could talk about it a little bit more. Is that weakness specific to Epsilon? Or are others seeing similar weakness? Obviously adjusting for maybe one or two large clients that have unique issues?

It is a good question. Don't know enough yet at this point. We do know -- I can only say what happened in our business, which was, across a number of our verticals, CPG and Telco and retail, it was interesting, we also have pulled back in Q1. It looks like it is beginning to trickle back in, as we start moving into Q2, but it still little bit soft.

So we don't want to call it yet, but it seemed like a lot of people just stepped on the brakes in Q1. So I just don't know what the others are seeing out there right now, Sanjay. But from us, it looks like Q1, people just pulled back. And then as we're looking at Q2, it looks like it is beginning to loosen up a little bit. And so if it continues to loosen up, we are not going to need to take any action. If it continues to stay soft like this, we are going to have to take some internal actions.

And I assume you are assuming continued softness for the rest of year in your guidance?

We are assuming, for the three-quarters of Epsilon and Conversant that's the tech and digital and everything else, we assume they're going to continue to be quite strong at about plus 8. And then on the agency side, to your point, we are assuming in our guidance that it will be soft, but not nearly as soft as Q1.

Got it. And than just second question, my final question. Just on those two portfolios that you chose -- or that are not renewing in card, could you just talk about how that unfolded? Why? And how tenured were those relationships?

Yes, this gets. It is a good question, and we are at the point now where we've got the business growing, the files growing 25% a year, which is pretty much -- we don't want to drift much higher than that. But we are also finding that the pipeline -- the number of opportunities that are out there, as people shift their dollars to the data-driven targeted spend, frankly, there's a whole bunch of business out there. And there are times in the past where we've had some clients who have decided not to pour all their energy and efforts into driving tender share growth and card growth.

That usually comes with a change in management. Other times, it's a change in strategy. And normally, we've just said, that's fine, we will go with the flow. Right now, we are dealing with, how do we allocate our capital? And if we want to continue to grow the business, the file 20% plus of the year, how do we maintain the type of returns that we expect? And frankly, there are a couple of programs there that based upon -- how do we say it -- the comments being made about the dollars that folks are willing to put into it, and the direction of those businesses, it did not look like the best use of our capital.

And so we have reallocated that capital to a couple of wins that we think are -- I think more down the fairway for us. So it is a nice -- I'm trying to be a polite way of saying, there is going to be some times where the capital is better used elsewhere, and this is one of those times. We've just never been capital constrained, but we are at the point now where the growth rate in the file is pretty much at our max level.

And how does it affect the growth assumption for that business? Does it?

It doesn't affect -- no, there's no effect at all. It's already been spoken for. And I know we had some -- already some questions about whether it was -- which client it was. And obviously, we haven't negotiated anything in terms of where we are going to put the file. But it is just FYI, it is not any client on the left coast. So that's all I can say on that one.

Okay, great. Thank you.

Darrin Peller, Barclays.

Thanks, guys. Let me just quickly follow-up on the Epsilon side. If you can give us a little more granularity, first into the strength of the tech and data side there? And I know Conversant, in the pipeline, looks strong still. But -- I guess I just want to make sure that's sustainable, in terms of, you're not seeing any of the same clients on the agency side pull back at all on the tech side. Obviously, the growth doesn't show that yet.

And then if you can give us a little more color on the levers that you have capable of pulling, from a cost standpoint, at Epsilon? And if it has to start next quarter, I guess how much room do we have? It just feels like an area that you've tried to cut costs in the past, and outsourced things to India, et cetera. So I'd love to just hear more color on how much room there is.

Yes, right now, what we are seeing is the continuation of strong demand for the digital products at Conversant, specifically the SKU level, data driven targeted display offerings. Very strong. In general, the more core Epsilon, big loyalty and customer database builds are strong, and data offerings are strong. So all the heavy tech stuff demand continues to be strong. We expect that to continue for the rest of year.

On the agency side, look, we need to be very careful we don't knee-jerk this thing. If it is an air pocket, then great, and we are back in business. If it looks like it is softer than we expect, then it is a very project-intensive, people-intensive business. You would have to move quickly to make sure that revenues and expenses sync up. But what we're seeing right now is, the softness in Q1 is already lessening a bit as we move into Q2. So we're going to hold off knee-jerking this thing.

And if, in fact, Q2 comes in very soft on the agency side, we will still make our [nums] for the Company, from a guidance perspective, for sure. But we will have to take some expenses out, to make sure we make our nums for the year.

Maybe I just missed it. But in terms of the 5%, what was your overall Epsilon outlook for the year now, in terms of organic growth, for the whole segment?

What we talked about for Epsilon is mid-single-digit plus on top and bottom. I think we'll stay there.

You're staying there now, okay. All right. Thanks. Let me just shift, and then I'll turn it back to the queue. But on the private label, on the card services side. I guess first on growth, do you agree -- co-branding, I know, is tougher now from a competitive standpoint. So just revisiting the addressable market, in terms of opportunities, obviously, given your divesting, I think this is the first time I remember hearing you guys divesting a couple of portfolios, based on better opportunities for probably more profitability elsewhere.

But either way, I guess just revisiting the addressable market would be helpful, given it seems like it is more private label focused than co-branding now. And then on that note, the yield compression is a combination of things, like you guys mentioned automated payments and new vintages, but also co-branding. With all of those things anniversarying and co-branding potentially moving lower now as a percentage of the mix, should the yield potentially rise in 2017? Or at least be stable?

I think the yield will be relatively stable in 2017, exactly to your point. As we are burning through these compression issues and the (technical difficulty) anniversary, and to your point exactly, a little less emphasis on the co-brand side, you won't have that large compression that you've seen the last couple of years of 100 basis points or so on the yield side. [It will be] significantly less than that.

And that has to do, again, with, we are looking to optimize the portfolio, which means that at this part of the cycle, you've got an awful lot of mouths to feed when it comes to co-brand. So that isn't quite as exciting to us at this point in the cycle. We've seen this play out before. So we are looking to optimize the files, continue the growth at 20% plus, and keep the compression much more modest on a go-forward basis. And based on the pipeline and what we are signing and what you will see released later this year, we feel pretty good with it.

All right, that's helpful, guys. Thank you.

Does that complete your question?

Yes.

Kevin McVeigh, Macquarie.

Great, thanks. Nice job, guys. In terms of -- Charles, the improvement in credit to 10 bps, what drove that? And how should we think about that over the course of the year? Still sticking with the [5.5%]? Or should we expect some improvement there, as well?

What we gave guidance to is a [5.0] loss rate for the year. (multiple speakers) We came in a little bit better in Q1. I think you will see we could come in a little bit better in Q2 and Q3, could be up a little bit in Q4 compared to original guidance. But I would say we are right on track to hit the 5% for the year. In terms of [missing] -- doing better by 10 bps of the first quarter, some of its timing. But I would say for the year, we're tracking right to the [5.0] loss rate that we gave as guidance.

Got it. And then as we think about the mix of private label versus co-brand, going forward, we are still in the range we talked about historically? Or more of a shift on the private-label side?

I would say we're still around 80/20, 80% private label, 20% co-brand. To Ed's point before, we're doing less co-brand. That's where most of the competition is, so we're focused more on private label. You know us. Private label is our secret sauce. That's where we have all the expertise. And so you will see more growth coming there, which, long-term, should benefit the yields, going back to Darrin's question before.

Awesome. Thanks. I will get back in queue.

Bob Napoli, William Blair.

Thank you, good morning. Hey, Ed and Charles, are the additional buybacks in your guidance? You bought back about 400 million. You said you had 500 million, and I think you expect to buy back closer to 800 million to 1 billion?

So the second piece of our authorization is not considered, Bob. So the initial 500 million was considered. The second 500 million is not considered in guidance.

Okay. Thank you. Then on the tender share improvement, the 80% from mature accounts, that is very helpful ago. But are you -- is that piece being driven by increasing credit lines for those individuals that are using the line? And obviously, since they're mature accounts, they -- I'm sure they are performing well, credit wise?

Yes. We are not popping up lines. We've never had, really, an issue with an open to buy. I think, Charles, you may know better than I. But if we have a $500 balance, we may have an $800 line. So there is plenty of room on the open to buy, and we re-score all 40 million accounts every quarter. So it is not a credit line issue. It is more of if I have an 800 line and a 500 balance, and I'm growing it normally 5% a year, how do I get it to 7% or 8%?

And that's basically the whole tender share game, which is, how do I get her to make one more visit if they are not visiting the store itself? A lot of our tender share gains, or I'd say the bulk of it, is coming through online sales, where we reach out to her and make some type of compelling offer to her, and she will go online to buy it. And a lot of this is, to use the lingo, she's snacking online. Meaning that we're seeing a lot of incremental purchases that are relatively small in size, but they are happening very frequently.

Great, thanks. And then just last question. On the agency business on the Epsilon business, is that -- have you lost -- is this just pullback by client? Have you lost some clients? Are you losing share of a client to another competitor or piece. So what's going on there? I don't think the industry has pulled back as much as we saw in your business in the first quarter. And who are you primarily competing with today in those sectors?

It is a good question. We don't -- one quarter doesn't a trend make, so it still up in the air, in terms of what exactly is [moving on]. All we did see was, we saw project business basically dry up in the first quarter that we had counted on. And as a result, the business we needed to close to make agency hit Q1 did not happen. Now, are those projects beginning to drift back in? It looks like they are slowly drifting back in. So that's all we know at this point.

And who are you seeing as your -- by sector, agency versus tech, who would you point out as your two or three largest and toughest competitors?

I think it is all over the board, in terms of when it is pure agency, you could have the big Madison Avenue players involved there. You could have some of the other digital smaller agencies involved there. But for the most part, what we do is, we like to link the agency as the tip of the arrow, to get in the door with the CMO to flow through to the big heavy-duty, high-visibility tech platforms that we build for them. The project-based stuff on the agency side, as you can tell, is the stuff that can jerk around quite a bit on a quarterly basis.

So that's what we are seeing. The big tech platforms continue to be built, and it is just these projects on the agency side that we just saw softness in Q1. So we just don't want to get out in front of this thing and start saying, my gosh, we've got long-term weakness in agency, when it could very well just be a one quarter phenomenon.

So Bob, the only thing I would add is, it's not a case where we're losing clients. It's not a number of clients going away, it's just less spend. So in the first quarter, we really saw less spend going back to its point where on a project basis, CMOs cut their budgets with us for whatever reason.

Great, thank you.

Andrew Jeffrey, SunTrust.

Hi, good morning, guys. Thanks for taking the question. Ed, I guess I want to ask a big pictures strategic question with regard to Epsilon. Is there -- as you think about agency, irrespective of the quarterly volatility in that business, is there still the same competitive strategic rationale for having agency as part of Epsilon as you might have thought there was five years ago? Has your thinking changed on that at all?

No. I think it is absolutely critical that we have that skill set, because that's the skill set that can get in the door, have the conversation with the CMO. And that will then lead to the pull-through to the big client database work, the big loyalty platform builds, the big CRM business at Conversant. So no, it's -- I think the skill set there is critical, in order to pull through for the other parts of Epsilon. It is almost like even though it is a bit soft this quarter, my view is, if you take it away, what you're going to wind up with is, you are going to cut the head off the rest of the big tech pull-throughs at the rest of Epsilon, and the whole thing will eventually slow down.

Okay. And is there -- with regard to, again, agency, is there a customer concentration issue, do you think, within that business? Is that part of the reason we see the volatility? And is there anything you can do about it if there is?

It is part of it, Andrew. If you look, our biggest clients in Epsilon are all on the agency side of the business. So when they pull back on any project or maybe not supporting a particular offering, they can move the needle quickly. And being project-based, they can do volatility intra-quarter quite a bit for the year. You can pretty well estimate what they are going to do.

But they can basically open up the faucet or close it quite quickly, based upon what they're wanting to do with their marketing spend. And so what we saw in Q1 was, they basically closed the spigot. We think that they will open it up as the year progresses, so we are not sounding alarm bells. But it is big companies, big budgets, and they can flex it quite quickly.

Okay. And then quickly, with regard to your outlook for what may be a loss rate plateau in 2018 and beyond. Can you dig a little bit into the rationale or the thought process that gives you confidence that we might see something that's pretty unprecedented, historically, in terms of stable charge-offs?

Yes, if you look at it, it is a great question. The fact that, as we talked about, the bulk of our growth, the 10% in the core growth that we're seeing, is really coming from accounts that are very mature, have been well seasoned, well past the peak charge-out cycle, we are normalizing the loss rate. I think maybe two more stats might be helpful, which is look, we thought, before the Great Recession, our normalized loss rate was 6.5%. We are saying it is 5.5% now. Why is that? Why are we comfortable that's our long-term run rate?

And that's what we are normalizing towards. Probably a couple of interesting stats there. One is, with the new mix of co-brand, which carries a higher-scoring customer, as well as some of the higher-end type retailers that have joined us, we are attracting card holders that tend to skew a little bit higher on the credit spectrum. So for example, if you were to compare all of our active accounts today versus 2007, the scores today, on average, are higher than they were in 2007, even with some of -- a lot of the accounts that we've added recently.

So from an active account basis, the average score is higher. Also, if you looked at it from a balance perspective, the average score per balance today is higher than it was in 2007. And so you put those two together, and that's how you get to, we are not going to normalize out at 6.5% -- we are going to normalize out of 5.5%. But look, it is not like we are trying to keep it lower. The fact is, you give up some yield by not being at a higher loss rate. So I'm not sure it is a strategy of ours to keep it at 5.5%. It's just the way the file is playing out right now.

Okay. Appreciate it, thank you.

Ashish Sabadra, Deutsche Bank.

Thanks. Just a quick follow-up on the agency weakness here. So the agency has (inaudible) worsened as well as core Epsilon. I'm just wondering if the 23% decline that you saw that was consistent across both those segments? And then Charles, you mentioned that there is intra-quarter volatility there, depending on the projects.

Is there something you can do strategically, going forward, to lower that volatility? Or is it just that as tech continues to grow faster, it will be a bigger portion of the revenues? I was just wondering, how should we think about the volatility coming down in the agency? Because it has been a volatile over the last several quarters now.

You are right, Ashish, we are going to have grow the technology piece of the business quicker. That's just frankly where it's going to be. Agency, in a good year, is going to grow 3%. Technology, in a good year, is going to grow 10% plus. So from a mix standpoint, you will basically take some of the volatility away from it. If you look at the breakdown, it was weakness in both categories. We saw weakness in traditional agency for Epsilon, where we saw we weakness across many verticals. I couldn't pinpoint it to any one client, because it was broad-based.

On the Conversant side, on the traditional display, we definitely have seen fewer people trying to use their private exchange. You've got the big agencies trying to push the programmatic buying through their own trading platforms. So regardless of the quality of the impressions we can provide, they're trying to stuff it through their own system. And we just saw a general pull-back there. So that was a little bit unexpected. So I would say it was both sides of it. We do think we will get some stability as the year goes on, but there's really, going back to Ed's point, nothing in particularly we can point to, aside from it just seemed like there was soft spend in Q1.

Okay. Just quickly, on the margins for the Epsilon business, you highlighted a couple of reasons for the margins being lower. But how should we think about margins, going forward? Technically, if agency tends to be a lower margin, tech was higher-margin business. And as tech is growing faster, you would expect the margins to go up, going forward. So just wanted to better understand, how should we think about the dynamics for Epsilon margins over the next several quarters, but also the longer-term?

So the way you would look at is basically, you are right. Agency will grower slower. It's the lower EBITDA margin. Your technology will grow quicker. It's the highest EBITDA margin, so just from a pure mix standpoint, you should get some EBITDA margin expansion in this year and in the future. Clearly, we will have to look at the agency and play through it. Is this the short-term hiccup or air pocket, as Ed said before? Or is it a case where we have to structurally cut back on cost to get our margins back where they need to be? That's a little bit of the unknown at this point. But long-term, I would expect the margins to expand slightly, based on mix.

Okay, that is very helpful. And one final question, on the gross [and compression]. So the late fees have been weighing on -- the lower late fees have been weighing on. Which is actually better, because it helps you -- or [it's reflective] of the better consumer environment. But just how should we think about that compression? How much visibility or how much confidence do you have that comes off in the second half and going into 2017? Thanks.

Yes, we have very good visibility into this, as we go into the second half. A large chunk of it, probably half of it, relates to what I call consumer-friendly programs, which I probably should define. It basically means we looked at all of our practices and basically said, how do we continue to be the premier provider in the marketplace, the most customer friendly provider in the marketplace, as we continue to ramp up? And so there's a bunch of things that we put in place, as we went into the back half of last year.

So for example, it doesn't seem like a lot, but it does move the needle, in terms of if someone gets paid on the 15th of each month, but their bill is due on the 13th, that doesn't work out all that well sometimes. So what we have done is, we've given everyone the right to move their due date to later on in the month if they wanted to. And that -- again, it doesn't seem like much, but it actually does affect the late fees. And -- but at the same time, it also helps our OpEx, as you see with the leverage, because we have much lower talk time of people calling up saying, I need a couple more days. So you have that.

You have people who are calling in, where they used to be a fee if you are paying by the phone. We decided that wasn't best practices. We took that out, as well. And so we just took a number of things out of the system that frankly caused more grief than they added. We got a lot of it back in lower OpEx, in terms of talk time and everything else, and we have a much happier client and customer base. And so I think that's what it is. None of it was dictated by any regulatory authority.

It was something we proactively went out and just did. And we think taking the hit now for it, and seeing it anniversary in the back half, is probably the best thing to do. And hopefully that will pay dividends down the line.

Thanks, Ed.

One more? Okay.

David Scharf, JMP Securities.

Good morning. Thanks for taking the questions. I'll apologize in advance, Ed, because it is a repeat of some that have been asked, but I just want a clarification. On the agency side, can you give us a sense for how much of the business for agency is so-called project-based? Versus, I guess, work that you would deem to be, while not an annuity, pretty reliable and recurring? Obviously, it is all based on CMO budgets. But is the vast majority of the business things that you would deem to be quarter-by-quarter project-based, where there isn't necessarily great visibility?

Definitely on the Epsilon side of it, that's the case, where it is -- you've got one or two year contracts, but it is project to project. On the Conversant side, it is going to more volume driven, because it's basically [site display]. Basically, you have got a lot of unpredictability when you get to agency. You can have a longer-term contract, but if it flexes based on volume, predicting what it's going to be on a quarter-to-quarter basis is just a difficult thing to do, David.

Got it. And when we think about the cost structure -- and obviously, the agency business has always had the least amount of operating leverage, given it's people intensive. But can you -- you've talked about, in the disclosure, that agency is roughly maybe one-quarter of the Epsilon segment, the traditional Aspen business plus Conversant. But from the standpoint of the percentage of fixed cost, how should we think about it?

I would say your premise is right, David, it is a higher fixed cost business. It's more of a people driven business, so variability is pretty low. Part of the reason why you saw that our EBITDA was down quite a bit in the quarter was because the revenue was down on agency, and it doesn't flex very much. So you are right, I don't have an exact percentage. But you can assume it is our highest fixed cost business would be within agency. So if we see it's a structural issue in the market, then we'll have to go through and address the fixed cost. And we can do so pretty quickly if we need to.

Okay. And as we think about the description of a lessening degree of revenue decline in Q2, 23% is a big number. Should we be thinking 15% to 20% down, or more closer to 10%? Just trying to get a little bit of context.

Yes, I don't think we are in a position to really say that at this point, as to what it is going to be. We do feel good that it's going to get better. We had a difficult comp in Q1, so we know it's going to be less than Q2. But at this stage, it would be very difficult for me to quantify it for Q2, Q3, Q4. As you know, David, we look at our guidance in total, not in any single piece. So just like in the first quarter we had over-performance in some areas and weakness in other, we would look for the year to be the same way. So it's very difficult for me to quantify that for you by quarter, David.

Got it. And as you referenced, a little bit of loosening in Q2, maybe some of the project-based expected or hoped for backlog is starting to materialize again. Are those coming specifically from those larger clients that may have pulled back unexpectedly? Or are these new wins, new contracts?

I would say it is a little bit of both. You do have some new wins coming through, and we'll look to focus on that during the course of the year. And I do think you're going to see budgets open up a little bit with us as the year progresses, as well. As Ed said, before, we think there was a little bit of a knee-jerk reaction with CMOs in Q1. I know I was told over and over we heading into a recession quite quickly in the US. So I think many people listen to the rhetoric, and I think on project-based, discretionary-based marketing, I think pulled back a little bit in Q1.

Got it. And then just lastly, on the card side. It sounded like some pretty guarded or crafted language regarding the two programs that won't be renewed. But is it fair to say -- should we think of that as two retailer partners that perhaps are not willing to commit as much marketing push behind the product as before? Basically deemphasizing the store card as one of their marketing tools?

Yes.

Got it. Okay, thanks very much, guys.

Okay. All right, everyone, thank you. We will see you next quarter.

Ladies and gentlemen, thank you for joining today's conference, and thank you for your participation. That does conclude the conference. You may now disconnect.