SLM Corp (SLM) 2018 Q3 法說會逐字稿

Good morning. Thank you for standing by, and welcome to the 2018 Q3 Sallie Mae Earnings Conference Call. (Operator Instructions) Thank you. Mr. Brian Cronin, Vice President of Investor Relations, you may begin your conference.

Thanks, Tarny. Good morning, and welcome to Sallie Mae's Third Quarter 2018 Earnings Call. With me today is Ray Quinlan, our CEO; and Steve McGarry our CFO. After the prepared remarks, we will open up the call for questions. Before we begin, keep in mind, our discussion will contain predictions, expectations and forward-looking statements. Actual results in the future may be materially different than those discussed here. This could be due to a variety of factors. Listeners should refer to the discussion of those factors on the company's Form 10-Q and other filings with the SEC. During this conference call, we will refer to non-GAAP measures we call our core earnings. A description of core earnings, a full reconciliation of GAAP measures and our GAAP results can be found in the Form 10-Q for the quarter ended September 30, 2018. This is posted, along with the earnings press release, on the investors page at salliemae.com. Thank you. I'll now turn the call over to Ray.

Thank you, Brian, and thank you all for calling in and for your interest in our company. It's a pleasure to report on the third quarter results, and I'll go through them in some highlights, and then we'll turn back to you all for questions. So it was a great quarter and it's a great year thus far with a good, terrific results in the market and on our margin and especially in risk. We're late in the credit cycle, as everybody knows, but our performance continues to improve. The sales growth in the private student lending business at 12.4% for the quarter greatly outpaced any estimate of the growth in the market, which is still typically thought of as 2% to 4%. We'll get the final numbers on that sometime during the fourth quarter, but our growth rate is multiples ahead of the market. And given we are head of the industry leader and have over 50% share in private student lending, it's all the more outstanding.

The other item, which is improving more rapidly than our expectation, is credit performance. Net charge-offs as a percentage of loans in the period were 88 basis points. The absolute number of write-offs are just about equal to last year on a receivable that is 18% larger. This, of course, is reflected in our EPS, which is $0.235 for the quarter, up 42% from the prior year, and our ROE 17% in the quarter, 19.6% year-to-date, outstanding numbers.

As I said, the volume in the quarter is a great story for us. And if you recall, we had estimated that we would be $4.8 billion-or-so last year, we thought that would go up a couple of $100 million. We doubled the rate of increase from our expectations, and it continues to hold, so that year-to-date we are up 10% over the prior year and this has grown through the quarters, so that was up 12.4% in the current quarter. This has all been done with stable credit quality and through a series of efforts that one segment to market more carefully than had been done in prior into core parent grad partner channels along with several others as well as concentrating on the actual click-by-click experience that our customers experience as they go through both in application and in customer service.

We now -- over 93% of our interactions with customers are done on the Internet, which is their preferred way of doing things without human interaction, and we have introduced chat as well as getting good results on our online application and mobile app. The customer satisfaction for chat remains at 97%, mobile application, outstanding, at 85%. And this all shows up in our market share, which is increasing, but done with a series of efficiency efforts, on one hand, greater segmentation, on the other, expanding our office with the 6 grad products and concentrating on multiple variables, no one single silver bullet as it were. In regard to that credit quality, the approved FICOs are absolutely flat at 747, the cosigned rate is flat at 89%, and so the increase in volume was not done while deprecating the quality of the credit portfolio.

Delinquency and charge-offs, the trends look terrific. And as I said, the receivable is up 18%, and gratifyingly the credit losses are flat to last year in absolute numbers. The NIM for the quarter at 6%, despite the fact that the third quarter is an area where, from a seasonality standpoint, NIM tends to be a little bit depressed, still up from last year's 5.85% to 0.6% (sic) [6%].

Operating expenses at $151 million in the quarter, there was 30% growth. Now this, of course, is higher than the ongoing growth and something that is not going to be sustained. It is exactly in line with what we told you, while we would do several quarters ago, so far as detailed investments and cloud migration, credit card setup, personal loan organic initiation. And this is 100% in keeping with everything we said, there is no new news on the operating expense basis. We continue to focus on the efficiency ratio, which was over 40% in '16, 39.6% in '17, and in '18 we're guiding to 38% to 39%, continuing our downward trend despite making investments in the company. And of course, we're not changing that because, as I said, no new information on the OpEx side. And as we look at '19, the efficiency ratio will continue to drift on down. So this is part of a 5- to 10-year story that we will have on an ongoing basis.

The credit performance continues -- last year, we had a 2.6% delinquency rate, this year we have 2.3%. And as you know, delinquency rate is a forecast model for what the losses will be over the next couple of quarters, and so the outlook is very good for us.

The balance sheet, meantime, is up 22% and the risk-based capital stays a very healthy 12.8%. The leaky bucket portion of this, which is consolidations, have remained flat, while the receivable and the paid portion of it has increased significantly. So if we look at consolidations for us in absolute numbers in the fourth quarter of '17, there were $243 million of consolidations. And as we went through the 3 quarters of '18, there were to $224 million, $221 million and now $228 million. As we said, plateauing there while the receivable is growing. So the consolidations are dropping as a proportion of our balance sheet. EPS continues its long-term trend. We've had 5 very good years there, and ROE reflects that. As a backdrop, our relationships in the regulatory environment continue to remain excellent. As with the FDIC, the UDFI as well as the CFPB, we have ongoing conversations with all 3. We work very closely with our colleagues there, and they have been nothing but supportive for all of our efforts.

In regard to the market frame of the customer, the competition and the evolution, in regard to that, the customer satisfaction, that we mentioned earlier, continues to play favorably for us. And as we've looked at our organic personal loan pieces, we're seeing that the Sallie Mae brand, on a variety of fronts, is much more powerful than we would have thought even 8 or 9 months ago. It gives us a competitive advantage in originating personal loans because the brand is: first, well-known; secondly, heavily associated with funding higher education; and third, well thought of by the people who have experience with us or with student loans more generally. We also remain a standout in the industry as an industry leader in thought process. We're working closely with new initiatives at the margin's remote colleges online learning, all of those things, along with our many partners who are, in many cases, at the forefront of the advancement of technology in regard to the market, but also in publishing select pieces, such as How America Pays For College and How America Values College, both of which have been well-received, highly circulated with millions of course contact points.

In regard to the employees, our turnover rate remains extremely low at 6.7%, our voluntary turnover, which, of course, accrues to a benefit of us, less hiring, fewer downtimes, less training, all of that. And so the evolution of market to segments is extremely important. The personal loan indicator so far as the receptivity of other products is very good. And, as you all know, the outlook has upped our EPS guidance to $1.02 to $1.03, the $5.2 billion originations we upped sometime during this third quarter, and the efficiency ratio is steady to our expectations at 38% to 39%.

So in closing this out, it's right to remind everyone, what a great market the private student lending business is. It's a unique niche that is filled by very few competitors. It is critical funding for a worthwhile cause by many of America's families. It results in our having a sought after customer base that has both initially very good performance with us and an ongoing basis great promise. Attractive returns are a commonplace, we're used to them, 19%, 17%, 18% ROEs, which are terrific, and the credit profile has been rock-solid. We, within that industry, are a unique asset. We're the best known name. We have market share over 50%. We have the largest sales force, which allows us to do the segmentation, as I mentioned. We have a modern digital platform, many partners that were diversified across the industry, and we have happily very conservative funding and very reliable funding. Our core earnings are high and the leverage is still building within the company, which you see in the efficiency ratios' 5-year decline. As we look out a little bit further, we have a strategic approach that is concentrated on allocation of capital and high ROEs for all the efforts in which we engage. You've seen us discontinue the purchase of assets nonoriginated by us because performance was not where we wanted it to be. We're ROE-driven. It didn't hit our targets. We stopped that effort. We will do the same with other efforts. We'll also evaluate all opportunities for ROE and the capital allocation that is derivative from that evaluation as we go forward.

And so in keeping with all the things that I mentioned, the performance that we have, it's a pleasure to talk to you about our employees' great work that has been done. It is -- we have a franchise that is unique, well managed. And I should say that in regard to these quarterly reports that we intend over years to do exactly what we said we would do. It is our effort to not have surprises in these things, to be hitting guidance, not to be routinely beating it because it should be a fair estimate, but to continue to deliver, deliver and deliver in regard to improving this terrific franchise, which we have the privilege of being able to manage. So thank you for your attention this morning, and we will move on to the next phase here.

Yes, I think, we'll open up the call for Q&A.

Yes, that's fine.

(Operator Instructions)

Okay, we're ready for our first question.

And your first question comes from the line of Moshe Orenbuch from Crédit Suisse.

Great. So congratulations on the really strong credit quality, and, Ray, you talked about that a fair bit. Maybe kind of help us understand how to think about the reserve development kind of as you go forward, the portfolio is going to continue to grow. It looks like you had a smaller reserve for troubled debt restructuring in this quarter. And maybe you could also kind of just clarify how we should think about the reserving on the personal loans. Any way to separate that for the stuff you've originated versus the stuff you purchased?

Sure, Moshe. And it's right to say that we essentially have 3 loan loss hitters that I'm going to ask Steve to comment on. One is the basic business, which is the CECL loan loss reserve, which as you can see the delinquency and write-off performance there continues to improve. And then there's the purchase personal loans as well as the organics. So, Steve, why don't -- I ask you to go through those, please.

Sure. Yes, good morning, Moshe. So as you pointed out, our TDR default rate has declined pretty substantially in both of the last 2 quarters. I think annualized, this year it was at 4% down from 6% in the prior year comparable quarters, so a pretty sharp drop there. And overall, defaults were actually lower on a dollar volume basis despite the fact that we had $2.75 billion more loans in full P&I. So credit is performing better than our expectations. I have asked the guys to take a look -- the guys and the gals to take a look at our life-of-loan TDR loss forecast because that is the basis on which we do our TDR forecast and the private student loan portfolio is performing very, very well. I think I've joked a couple of times on the call that we keep targeting on 1.5% reserve for the private student loan, and we never quite get there. And that was the forecast, I think, I gave last quarter. And thankfully, we are struggling to meet it. On the personal loan product, the vast majority of the bill this quarter was for our purchase portfolio. As Ray has pointed out, we are not meeting our return targets there. And one of the primary drivers of that is that losses on that portfolio are exceeding our expectations. If you bifurcate the reserve for the personal loan portfolio, this quarter, it's running about 5% for the purchase portfolio, 2.5% for the organic portfolio. We believe that our personal loan portfolio should continue to perform better -- I'm sorry, our organic portfolio should continue or will perform better than our purchase portfolio for the most part because we're dealing only with preapproved credit. So in essence, we underwrite before we offer the product to our clients, which is considerably different than some of the marketplace lenders out there. As we move forward, the reserve for the purchase portfolio could grow a little bit more. The organic portfolio reserve should approach 5% as that portfolio grows and seasons. So I think I covered all the questions that you had, Moshe.

Yes. Perfect. And maybe just to follow-up on the expense front, you talked a little bit about -- could you just maybe describe whether the Q3 expenses are impacted by the level of originations that you had, which, obviously, was higher than we were expecting? And also, you've mentioned you spent $19 million of the $40 million that you had earmarked for your kind of ongoing strategic efforts. Does the rest of it gets spent, how do we think about that?

Sure. So on the core student loan business, the expenses weren't really driven by the increase in originations. We're still seeing flat to slightly down applications. We continue to invest in our platform there. So the 14% year-over-year growth in OpEx in the core student loan business, a big chunk of that was from growth in units, but there were additional expenses associated with things like CECL, facilities expansion. For example, our call center and collection centers are growing as are our loans in full P&I grow. And there are a couple of other projects that we continue to spend money. Ray mentioned, chat, for instance, which is one of our newer initiatives. On the expense side, on the $40 million that we've carved out, I think, it's important for people to understand there that except the $10 million -- excluding the $10 million that we are investing to migrate our IT infrastructure into the cloud, the remaining $30 million will continue in 2019 and will be driven by the amount of personal loans that we target to originate. I think we've talked about keeping our volume flat at $500 million into 2019. And it'll also be driven by progress on our credit card initiative, which will begin really in earnest in the -- in 2019, later in the second quarter and second half of the year. So I think, I covered your questions there.

Your next question comes from the line of Sanjay Sakhrani from KWB (sic) [KBW].

(technical difficulty)

Sanjay, we seem to have lost you. Operator, are you there?

I am. (Operator Instructions)

Hey, can you hear me guys?

Yes.

So to what do you attribute the faster-than-industry origination growth? Is it about the strides that you've made, or has something happened in the competitive environment as well?

No. We actually -- addressing your last comment first, we don't think the competitive environment is volatile -- maybe it's pretty much the same competitors, it's the offerings we look at very closely. And so in the private student lending market, I would say that has not been a factor. I do think that we are aggressively pursuing many facets of this, and I believe what we're seeing is the benefit of that. The first is, of course, the ongoing improvement in our operating platform. Our customer satisfaction numbers continue to improve and are over 90% on our online application, which was new about 18 months ago. And within that, we also look at each stage for pull through. We also look at, in much more detail, what people are borrowing money for. And so we walk, in some sense, electronically through the process that our customers, the students as well as their families are going through as they think about how they will finance higher education. Now we are more involved in that in the application process now than we have ever been. And that has resulted in, one, higher satisfaction and, two, higher volume for us. In addition to that on what had been treated largely as a homogeneous market is clearly not one. And the grad offerings, of which we've introduced 6, are full slate, targeted to the various professions, as you might imagine, medical, legal, MBA, that sort of thing, and we've received good results there. Our grad volume is up about 18% year-on-year. In addition, we've made a commitment to wrap ourselves around partners in a way that's good for them and good for us. And so our partner business, which is a small portion of things overall, but as I said, if we look at 1% here, 2% there, it adds up over a period of time. Partner business is growing much faster than the market, good for our partners, good for us. And of course, it forestalled the possibility that any of those partners would turn into competitors if they're seeking to participate here. And so it is with sort of a scalpel around segmentation, around customer experience and continuing good results with the colleges that has resulted in the outperformance here, which is across the board. But as we said, we don't think about this anymore as one lump sum $5 billion. We're working now on about 5 or 6 active segments with varied solutions, which give better satisfaction on both ends of that equation. So an answer to your question, no big change in the market, but with a series of improvements and product offerings, customer satisfaction, pull through and then just working with families as to how they expect to finance their children's higher education, we received this year gratifying results, and way above our expectations that we would have had in January.

Absolutely. And then on expenses, those were up about 30% in the quarter if you exclude the investments more like mid- to high-teens. Is that a better growth rate going forward?

So look, expenses are largely driven by units in full principal and interest repayment. This quarter is a tic slightly above that growth rate. And we focus an awful lot on our efficiency ratio, and we are not going to make improvements in that efficiency ratio, if that continues to be the case. It just so happened that in this quarter, we did have some unavoidable expenses, such as the CECL investment that I called out, but we will work to get that growth rate and OpEx in the core business back to the growth rate of units involved in the portfolio.

One last one. Steve, you mentioned the personal loan portfolio losses coming in ahead of your expectations. Could you just elaborate on that, like what was driving that? Is it the third party -- the loans that you selected from the third party or how did that work actually?

Well, so I want to make sure I understand your question, but I'll repeat what I think I said. So in our organic portfolio, we look at performance of loans month on book delinquency rates at this point because we don't have a whole lot of loans in the portfolio that are approaching the point at which they start to default. I think we did have something like 7 defaults, but the delinquency rates on that portfolio are trending significantly lower than the delinquency rates at that stage of the seasoned -- of the portfolio than our purchase portfolio performed. So I was trying to make the point that we think our organic portfolio will perform significantly better than the business we just terminated, and we should be able to meet our return hurdles in that product.

But I think, Sanjay, your other point was, was there something about the purchased assets that introduced a dynamic here, increasing the loan losses associated with them? And I believe the answer to that is, no. The quality of assets that we're purchasing has been pretty constant, but the expectations of the originators for their losses in a constant segment quality purchase scenario are actually higher. So it's a performance that deteriorated, not the purchase screens.

Okay. And do you guys feel comfortable that it won't deteriorate a whole lot more going forward or -- because we're past the seasoning cycle or...

Yes. Look, we're going to make, by many people's standards, a very acceptable return on these assets. We just, given the nature of our business and the alternatives, hold ourselves to a pretty high hurdle rate. So we will not lose money on this portfolio. We will just not make the returns that we think are appropriate, given our cost of capital, et cetera, et cetera.

Your next question comes from the line of Arren Cyganovich of Citi.

Steve, you had mentioned that the application volumes were kind of roughly the same. You had really strong origination growth, 2%. You talked about the ticket size going higher. Are you also approving a higher rate? What's the difference there?

So there is no change in our credit box. Our credit standards have remained absolutely the same. It is the case that we are seeing better pull through, overall, in the application process. And as Ray, I think, explained in some detail, we are targeting more effectively different segments of our portfolio and growing. For example, to the point you made, our graduate student loan business grew at a rate of 18% this quarter. So we rolled out those 6 new products. So you put it all together and the business is firing on all cylinders at this point in time.

Okay. And you had mentioned some partners that you have that you're working with. I now recall you guys talking about that much in the past. It is just banks that previously were involved in the business and now are referring volume to you, how do those relationships work?

Sure. And it's a full range of partner profiles. But in general, you may think of smaller banks, credit unions, people who have a customer base and they would like to provide the benefits for a private student loan, but they don't have the leverage to develop the platform that we have nor the compliance opportunities or larger compliance requirements that are exigents for it. And so it's a long list, I believe the number is over 750 at this particular juncture, of small players that we partner with. It's good to them, it's good for their customers and good for us. But largely people who have a financial services platform and don't have, as I said, the volume that would warrant, they're putting together a separate platform for private student lending and we partner with them. It's a number that has -- we've been concentrating on it over the last year and the growth rate in the partner channel is coincidentally 18%, as Steve mentioned for the grad piece.

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

Many of my questions have been asked. I guess, just a couple of kind of nuanced questions are for the originated unsecured loans, couple kind of characteristics I'm interested in is, number one, for your originated loans, what is the average size and rate? And then number two, you mentioned some of the kind of channels to get the customers, but I'm wondering how much of them are cross-sells from customers that you either have or had private student loans with versus new customers altogether?

Sure. And so first things first. This is 2018, we launched the product in January. We're doing a bunch of testing in 2018. And so we're looking at people who had some association with Sallie Mae at one time, no longer do, people who are currently customers, people who have had some familiarity with student lending more generally and others who are creditworthy. And so as we look at that, the cross-sell portion of that, our customer base is not overly large. It runs depending on how you calculate between $1 million and $1.5 million. And so the cross-sell opportunities are now much better than, of course, a random sample or a targeted nonrelationship population will be a small portion of things going forward. It is the case though that the general demographic that we're interested in, which is young people and how they sort of get started in life, of which personal loans are a piece, will be a very important part of the originations. And so right now we're testing across work. And I have to say, as I alluded to earlier, that the Sallie Mae brand is both unique as well as powerful in regard to these tests. And the additional pull associated with our brand versus either anonymous or random for people who have had some association with funding higher education is terrifically in excess of anything we thought and significantly lowers our origination cost for those large populations, which runs in the tens of millions. And so the brand actually allows us to have a proxy for cross-sell that others might have with a preexisting product because we're uniquely associated with higher education and, in fact, dominant within the private student lending piece. So the average ticket is running about $17,000. The range of APRs is a full range from 5 to 15-or-so. And as so we're testing on this, and as we enter 2019, we'll be taking the results of those tests more towards the profile that we'd like to have for this portfolio.

Okay, that's helpful detail. And then second question just sort of thinking about kind of intermediate term trends. Where -- how much is the P&I portion of total loans in full P&I now? Is that going to be balanced with where it is a year from now? And then depending on that change, what does that mean for things like delinquencies and charge-offs? How do we think about the effects of that?

So John, we're running around 36% full P&I, and what happens and hopefully it will continue to happen. We continue to originate large volumes of loans, so that number tends to rise and fall as our loan origination volume increases. It is on a very gentle trend higher. But it is the case that we have -- I want to say, it's $7.5 billion in full P&I. And for those of you who have been with us on this journey, we started to have it right around $1 billion. So we feel very good that having multiplied that by 7.5x, our credit performance is not only meeting, but beating expectations. It could very well be the case that a big chunk of this portfolio is past its peak default rates. So if the portfolio was to stabilize in size, we would see defaults trend lower not higher, but I think what we're going to see is that they're going to stay, remain stable around this 2% level of full P&I going forward.

(Operator Instructions) And your next question comes from the line of Vincent Caintic from Stephens.

Just want to take maybe a step back and talk about the EPS guidance. So you've had a nice raise to the 2018 EPS, and it's now -- that's indicating above 40% year-over-year EPS growth. Just kind of wondering maybe if you could describe what's driving you to increase the 2018 EPS guide? And then as you're thinking about 2019 and the different components of originations, naming the credit, is that something that you expect trends to continue and maybe not 40% year-over-year, but to get to a double-digit EPS growth?

So the key contributor to us raising guidance for the final quarter year-over-year was credit 2/3, NIM 1/3. We do expect the credit performance to continue to be rock-solid. I bumped into one of our credit guys yesterday and I said "Hey, is it time to start second-guessing our life of loan default rates?" And he reminded me that we are right smack in the middle of one of the strongest economic environments we've seen in decades. So he wasn't in a big hurry to do that. But we're at 3.7% unemployment and as long as those trends remain solid, I think we can expect the portfolio to do very well. But you know credit is very much economy-driven. There are no signs if there's any major deterioration on the horizon. So going into '19, if things are looking similar to the way they are today, we can expect that, that will definitely be in the plus column as far as EPS in 2019 is concerned. NIM, look, we -- we're in a situation today where we now have -- I think, it's 32% of our portfolio's fixed rate. That's up substantially from a year ago when it was 22%. 65% of our loan originations are now fixed-rate. So that's a much different game that we're playing today than we were a couple of years ago when interest rates were not moving. Fixed-rate loans are somewhat more challenging to fund, but we think we are appropriately positioned to maintain the 6% ZIP Code NIM that we are going to put up this year and in subsequent years. We do take conservative approach to funding, and we try to lock that NIM in, not take risks that will potentially increase or decrease at the margin.

Okay. Great. And then just one quick one. When you think about the credit card portfolio 2019 rollout, any sense for how big of an impact that might be? And is there -- when you think about incremental expenses or anything like that, your credit and so forth, what does that add to the portfolio?

So on the grand scheme of things, it will be 0 credit card receivables on the balance sheet at the end of 2019. There will be expenses associated with continuing to build the platform out. And credit card is very much a direct-to-consumer marketing game. So to the extent that we are ramping up our marketing machine in 2019, there will be expenses associated with that. I think we're going to have to save that discussion for our January conference call, where we give you the finer points on our diversification efforts, but we are very much committed to not harming the core earnings, the earnings generated by the core business. Ray has said in public that we would never risk more than 5% of those earnings on any diversification effort. It will be the case, as we look out to 2019, that the personal loan will no longer be a drag on earnings. I'm not telling you that we're going to spend $0.05 on credit card because I think that, that would be way off the mark, but we're committed to continuing to generate strong earnings growth at Sallie Mae here.

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

Steve, I just sort of have the same question for you that you have for your credit guys, which is, a year ago you talked about the idea that lifetime losses on the portfolio -- on a cohort would be about 8%. You talked about vintages running towards 6%. What I'm curious about is, when do you see sort of peak losses in a vintage with the idea that for an individual cohort, you could actually take down that cumulative loss expectation?

So peak losses are behind us really by about 30 months into full P&I. And in the future state of CECIL, I think, analysts are going to love some of the byproducts. We will actually be publishing expected remaining defaults by cohort. But we have, in any given year, $2 billion, $2.5 billion of loans going into repayment and big chunks of our $7.5 billion in full P&I are past that peak default rate.

There are a couple of points here, right? One is, our loss credit follows similar other consumer lending products, right? So typical P&I runs the, as you know, for a life of loan loss about 7 years. And so in the first 2 years of P&I, you get about 50% of the losses, right? So that's 28% of the clock time, 2 out of 7, and it's 50% of the losses. So it's a 2x multiple for the first 2 years. Then after that, things trend on down. But it's right to say, as Steve has alluded, that our credit performance in 2018 has exceeded our expectations significantly. And so as he alluded to with his hallway conversation for -- with our credit manager, it is the case that we have unexpected good news, we look into the future, people say, "Oh, gee, late in the cycle to worry about risk embedded in the portfolio." We have all the data that we need from the last 3 sessions. We know that our portfolio is less volatile than many of the other consumer portfolios because of the high quality of the families to whom we lend. And so what you're hearing is a good business, which has great returns, unexpectedly good credit performance and starting to think, gee, is this unexpectedly good credit performance a new norm in such a way that we could leverage it either with lower pricing or higher LIBOR rates, given that forecast, or should we remain cautious given where we are in the economic cycle and in newness of this information? So we're at a fork in the road, which says, do you want to continue as you were doing, which is just fine, or could you actually improve your full year forecast for the losses as well as the NPV for newly originated accounts, and we're debating that right now. But it's an opportunity quadrant to be, not anything else.

So it's a very interesting way of looking at it, and I share that yield. If we start with the idea that it is not a new normal, but rather that there are certain vintages that have benefited from historically low unemployment and that those vintages will produce lower cumulative loss rates. Which of the vintages do you think at this point are seasoned enough that you could actually make that assumption that due to macroeconomic factors they're going to experience -- those cohorts will experience lower losses?

Well, if we take that and just plot it on a chronology graph, right, we'd say, on average, if we made $5.2 billion worth of loans in this calendar year, when would they go into P&I? On average, they will go in about 2.5 years after that, right? So the 1/3 entering full P&I today, third quarter of 2018, are those people who on average had loans that were in '15 and '16. And so now plot the unemployment dropping for the macro environment against that, and you would say that, "gee, people who are in full P&I today in fact were only benefiting from the beginning of this very favorable employment cycle" because, as we all know, it has gotten rapidly better over the last 2 years. Almost none of the originated loans in the last 2 years are in fact in P&I. So that improvement that you're anticipating, which is I graduate and get a job at a higher level than the prior cohorts, is still in front of us.

Your next question comes from the line of Michael Tarkan with Compass Point.

Just a couple follow ups. On the yield question, can you tell us where yields are on the new originations relative to the -- average yield on the originations you put up this quarter relative to the 9.16% for the overall book?

Sure. So the loans that we were purchasing had an average coupon around 10%. We are a 0 fee lender in the personal loan business, and our coupons are going to be approaching 15%. They're not quite there yet, but they will be more yielding than what is currently on the books.

Okay. And then what about on the student loan side? Just wondering where average yields are for the peak lending season?

Sure. So on the student loan side, our average yield on the variable portfolio was 7.5% over LIBOR, so basically 10%. And the average yield on the fixed-rate portfolio was 9.75%. Very flat yield curve, hence people selecting that fixed rate in our current rising-rate environment. And while I have the microphone, I stated earlier that we're going to have no credit card receivables on the books at the end of 2019. Apparently, Brian, got a few e-mails on that from the team. I standby, my number, it's going to be less than $50 million. So on the grand scheme of things, as analysts are concerned, there will no receivables on the book. I am sorry, Michael, you had another question?

Yes. Just on the personal loans, the cycles that you are seeing come through the organic product versus the purchase product, are those largely comparable?

They are very similar. But of course, we are using our own custom credit scorecard. So FICO is not the only variable that we will be looking at. And I did mention earlier that these are preapproved credits. So we do think for a number of reasons, they will perform better than what we currently have on our books.

And then just one quick one on that one. So are you still sourcing primarily those borrowers from your existing student loan pool? And at what point, do you broaden that out as you look to build that product a little bit more?

Yes. So we've already broadened that out. As Ray mentioned earlier, our current client base is just over 1 million borrowers. So we could not build a personal loan business on our portfolio alone. So we are purchasing lists that consist of people very similar to our current borrower list, and it is a little bit more broad as well.

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

Just real quick, the credit did come in better than our expectations as well. And then could you just talk about any adjustments that you guys have made at SLM over the last year that may have contributed to some of this outside of just good economic conditions?

As we've reported quarter-by-quarter, we've been quite consistent in our credit approach, both in the cut-offs of through-the-door population, the cobranding proportions, where they come from so far as to schools. And the improvements that we've tried to implement, and we've had success in this, is the biggest issue in this type of credit, we've talked a little about, who is in full P&I, who is not in full P&I. As it turns out from a credit-bubble standpoint, the biggest concern that we have in a business, such as ours where there is deferred payments from many customers, is the transition from state A to state B, either in a minimum payment, just paying interest or some token amount, and I go to full P&I or I have full deferment and I go to full P&I. And that mini wave is the most difficult thing to forecast. It is most difficult thing to staff against. And we have done everything we can to mitigate that wave height. And so we are now sending multiple notices to people, gee, in 4 months here, you're going to have to make a payment here, in 3 months, you are going to have to do this, asking people whether or not they are ready to contact us as a priority if, in fact, they do have some difficulties. And we have been quite successful in getting people going from 0 to 1, not to do that. And lowering that initial wave, of course, has a benefit in our staffing, our delinquency and, ultimately, in our losses because if people get behind, catch up is really difficult. And so this is an operational type of customer approach, which calls for a partnership between our customer service units and our collection units, which is greatly enhanced over something, which didn't really exist in that form 3 years ago. So no change on the front end, and many improvements on the back end. But as Steve said, the economic environment is also crucial here. And while our portfolio in the prior recessions has had a much slower beta than other consumer portfolios because of the high quality and because of the family lending and the cobranding, it's still the case that we're definitely benefiting, as are others, from the benign environment.

Great. Excellent. And then quickly because you guys have done a great job stemming the growth in loan consolidation to third parties, you've talked about some of the strategy around this in the past, but when you think about reducing the rate on the loan versus extending the term to lower the payment once someone graduates, how does -- what's the thought process around that? And where is the cutoff there when you say, you know what, this isn't economic to us, we're going to let this go as well as not maybe lowering the rate or extending the terms to the point where somebody comes -- the same person may come back for another cut, maybe 6 months later, what's your thought process there? And where do we -- where can we maybe see going forward more based on extending the term or lowering the rates? Where do you see all these things moving?

Yes. As we've talked about in prior quarters, we've done relatively little in terms of volume, either in lowering the rate or extending the terms. And the customers, of course, have a focus. If you survey them, they will say, "I want to have a lower rate," but if you ask them how you think that will show up, they really need a lower payment and so the lower rate is a means to an end. It's also the case that in the customers we've looked at, they -- when they're consolidating, they're not out consolidating Sallie Mae loans alone. They are consolidating their credit card business, other loans they have. The objective is a better cash flow for their households. And so we're continuing to test in regard to this. But as Steve has said on multiple quarterly calls, the backdrop interest rates that have gone up consistently over the last 1.5 years, they have mitigated some other competitor's enthusiasm for this business. So we are going into this in a small way. We have some tests planned for 2019. We will continue to build our capability to engage in this type of fighting. We do think it is, on its face, a margin destroyer. So our enthusiasm for this type of business is relatively low. The volume, as you've seen, has plateaued almost exactly from the fourth quarter of last year, it's actually lower than the fourth quarter of last year, to where we are. And so we're arming ourselves to be ready in the event that this escalates, but we would prefer to keep a low profile in this. And if the benign trends that we've experienced over the last 4 quarters continue, we will operate defensively, but not dramatically.

And your next question comes from the line of Michael Kaye of Wells Fargo.

I was hoping you could talk a little bit about your recent investment in Deserve, and how that fits into your credit card strategy?

Sure. Deserve is a west coast, very modern, integrated native app purveyor of credit cards. And when we looked at our entry into the card business, several things guided us: One is that the card business is filled with very capable competitors, many of whom have excess capacity, while we're sitting here; two is that we did not want to build an infrastructure that had high fixed costs associated with it; three is that we wanted to be modern and we thought that we would have an advantage over existing players who frequently are bound by their old unintegrated and not fair consumer-friendly systems. And so as we surveyed who would be a potential partner, we looked at multiple potential capabilities. We hit upon the Deserve folks because they are modern, they are dedicated to us and -- but they are a relatively new company, and so we made a small investment in them. We have observer rights on their board. We are working closely with them. And as Steve has alluded to, we're expecting to have a friends and family type of walk-through for our product within about 45 days of today, a second chapter of that in January, and we expect to be testing our way into the markets in the second quarter of next year, which we signed the contract with Deserve in second quarter this year. And if we can move from a contract signing to a product that is working in less than 6 months, I believe that will be a new standard in the industry where that lead time is typically measured even for the best competitors in 12 -- as 12 to 18 months. So the responsiveness is good, they're very modern, they do have a native app, which makes for an integrated customer experience, and so far so good with Deserve. But a small investment along the way was called for just because of the fact that they are newish.

My second question was, I just wanted to get some thoughts on the upcoming mid-term elections. Let's look at a couple scenarios. If the Republicans lose control of Congress, how much of a setback would that be for something positive developing on some private student loan expansion? And if the Democrats do gain control, would that just really perpetuate the status quo? Or is there risk of something negative developing?

I know this isn't a political discussion, but when I look at -- think about your first comment, if the Republicans were to lose control of Congress, one might, and two if not, now you think they do have control of Congress. And I think there would be different opinions on that even within the Republican Party. But -- while we've seen over the last -- we're coming up on our 5th -- December 31st since the spin, and we've spent quite a bit of time on the political front. And there are people with extreme dispersion, everything from get rid of the federal program to college should be free for everyone, as you know. And what we've seen thus far, over the last 4 years, it's basically gridlock. We have -- the Higher Education Act has not been renewed, the initiatives have died almost the day that they are announced. Of course, the deficit is improved -- has not improved, rather has increased. The federal student lending business at $1.6 trillion is a portfolio that no one in Washington seems to want to address at all. The losses embedded in that we know are approaching $0.5 trillion. And so I would have to say that if you were just forecasting this coldly as opposed to what I might desire or something along those lines, you're going in bed would have to be more gridlocked because even on the Democratic side if you look at the people who would be in the majority in a blue wave house of majority change, the student lending is really just not in their top list. And so we don't really see any formula plans there. People have said, all in the end, the government would have to get out of Grad PLUS or to get out of Parent PLUS and there are people for that. That would greatly benefit our business, no doubt about it. On the other hand, raising the limits for the undergraduate federal loans would hurt it. And so we've run all those scenarios, but so far in the course of 4 years, there has been no change. And so from the standpoint of empirical feedback should make us humble, I would have to say going in thought would be no change.

And there are no further questions.

Okay. And so I want to thank you all for your time and attention. And as I said earlier, it's a pleasure to report these very good results. Steve and I were talking about what would happen on this call before the call, and I think if we had a catchphrase for the third quarter of 2018, it would be as we told you, the following: The results are right in line with what we thought with 3 exceptions: One is the purchased personal loan business has not performed the way we liked, and I think discontinuing it is, of course, the right decision, in my opinion. But I think it's an illustration to our analysts as well as to our shareholders that we are willing to be bound by ROEs and when they're not sufficient, as Steve said, they're still positive, but not sufficient, we will act in keeping with capital allocation that is done rationally. The other 2 items are the origination volumes, and my vantage point is from January is significantly higher than we thought, seeing a payback for our multiple leverage, which is extremely gratifying, we will end the year as the industry leader starting with the market share over 50%, up 10%. So that's fivefold share points. In a market that we think is growing 2% to 4%, we think some of our efforts have actually expanded the market volume, and we can talk more about that in the future. And as we've talked about significantly in this call, our credit performance is doing much, much better than we thought, and we expect that to continue for the foreseeable.

So in summary, this is a great and unique franchise. It works with an upscale customer base doing things that are very important to them, which result in a high customer satisfaction, which we now have at 80% overall, and excellent performance on credit. I should note that the management team strives to do exactly what we told you we would do. We're in a no-surprises business, we look to hit guidance, we look for the guidance to be a fair estimate, not a lowball piece, and we are in the business of deliver, deliver, deliver. We have a great franchise. We will address it as best we can. We will do capital allocation in a rational basis as we go forward. And I thank you all for your attention today.

Great. Thank you for your time and your questions today. A replay of this call and the presentation will be available on the investor page at salliemae.com. If you have any further questions, feel free to contact me directly. This concludes today's call.

Again, this does conclude today's conference call. You may now disconnect your lines.