SLM Corp (SLM) 2017 Q3 法說會逐字稿

Good morning. My name is Angel, and I will be your conference operator today. At this time, I would like to welcome everyone to the Sallie Mae Third Quarter 2017 Earnings Call. (Operator Instructions) Thank you.

I would now like to turn the call over to our host, Mr. Brian Cronin, Vice President of Investor Relations. Sir, you may begin your conference.

Great. Thank you. Good morning, and welcome to Sallie Mae's Third Quarter 2017 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 discussions 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 to GAAP measures and our GAAP results can be found in the Form 10-Q for the quarter ended September 30, 2017. 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.

Okay. Thank you, Brian, and thank you all for your interest this morning. It's a pleasure to be able to report our third quarter results and talk about where we're going from here. I'll walk through a series of quantitative measures with some prose comments, and then I'll turn the call over to Steve, and then we'll open for questions.

So in the quarter, we originated $1.9 billion of new private student loans, up from $1.8 billion last year. The growth rate was 3%. That also is the growth rate that we have year-to-date, where we've originated $4,166,000,000 new student loans.

As we look at this, we're actually very interested in the trends within the market, and the latest data that we have, which was for the first half of 2017, provided to us by Measure One, indicated that the market had been growing about 1.27% up to that point.

For the size of the market, we believe that, that growth rate has continued, and applying that to the overall market, that would be an increase of approximately $90 million in originations.

As we look to the full year, our volume increase will, of course, exceed that. So not only will we match the market's growth but we will also be increasing market share, as we have done over the last couple of years. We'll talk more about that as we go along.

We naturally are very interested in credit quality through the door. It gratifies to say that, that has been 100% consistent. We had a 747 average FICO score in the third quarter, approximately equal to last year's 749, and we had 89% cosigners versus 90% last year.

We continue to monitor our take rate. After we've approved customers, do they in fact take the loan out with us. We're happy to report that we have that -- the take rate has been consistent at approximately 80%, just a hair over it.

So then we look necessarily to NIM. How efficient is the asset? Our NIM in the quarter was 5.85%, up from 5.58% last year, a 27 basis point increase. We are, of course, in a mild interest rate increase era, and we are experiencing, as we have said in prior quarters, a gratifyingly lower beta on our consumer deposits than we had originally anticipated. And as you all know, we thought originally that the NIM would be consistent through the cycle. But to the extent we're seeing some improvement, it is very good.

Operating expenses were $116 million in the quarter versus $100 million last year, an increase of 16%. As we look at whether or not that's a reasonable number, we track it against the increase in the receivable, which of course, is how we spend most of our money for servicing.

So private student loans at the end of the quarter were $17,186,000,000, up from $13,889,000,000 the prior year. So that increase of $3.3 billion is very good in absolute terms, but it is 24%. So we like the receivable up 24%, operating expenses up 16%, and this allows us to continue to be confident that over a period of time, not only will we deliver excellent efficiency ratio numbers but they will continue to improve.

This is all being done while our customer service is improving. On our customer ease index, which is do customers interact with us in the area and in the methodology that they choose, we are now experiencing a continuing increase in that number. So customer ease index for us is 92%, which means that of all the times that a customer would interact with us, 92% of the time, they self-service, and only 8% of the time are they required to speak to a human in order to satisfy a more complicated problem.

So the investment continues to drive enhancement for us, the volume continues to grow, the efficiency ratio continues to be attractive and it's trending in the way we anticipated.

In regard to credit performance, losses in the quarter were 1.08%, up from 91 basis points the prior year. These losses are on our models. As you know, we forecast life-of-loan losses when setting our ROE objectives as well as setting our cut-offs for risk.

Delinquency in the quarter ran 2.26%, up from 2.20%. We looked at this very carefully. We're quite conscious of the fact that in the industry in general, delinquencies have increased, losses have increased and there's been a particular focus on the credit card arena in regard to that phenomenon.

As we look at the 2.2% versus 2.26%, the question for us is, gee, is that something that is an aberration, or is this a trend about which we should be concerned?

But first, we return to our model. The model would indicate that this 2.20% should be about 2.3% this particular quarter. So that would be sort of the baseline from which we'd compare things. We do believe that there was about 10 basis points, on average, associated with the various weather conditions that were troubling some of our customers along the -- especially in Texas and Florida during the quarter. So we got 2.2% or 2.2% -- we've 2.3% for maturation, 2.4% for flooding. And then over the summer, as we tried to address seasonality and staffing, we think that we were a little bit slow on some of these early delinquency buckets. We have since corrected that.

As we sit here in mid-October, this delta of 60 basis points has entirely receded. And so the 2.2% jumps to 2.26%, I'm going to call that a jump, increases to, and it looks as though it is transient. As I said, we're concerned about this because of the results that others in related spaces to ours, consumer lending, especially unsecured, have experienced. We're quite confident that we are on our model. The numbers grew a little bit, mostly associated with weather and seasonality, and that bulge has been entirely depleted, as we speak.

Our balance sheet. Balance sheet is at $21,016,000,000, up from $17.7 billion last year. That's an increase of $3,270,000,000.

Our private student loans, as noted a couple of minutes ago, are up $3,297,000,000, which means more than 100% of the growth in our balance sheet is being provided by productive assets, which allows us to have, obviously, the better NIM, better returns and, of course, in general, a more efficient balance sheet.

The results of all of this is that EPS in the quarter was $0.17, up from $0.12 the prior year, a 42% increase. ROE at 14.6%, up from 12.2%, a 20% increase rate-on-rate.

Turning to the environment for a minute. The regulatory relations that we have with the FDIC, the Utah Department of Financial Institutions and the CFPB have all been excellent. The FDIC and ourselves in the field activities have been working through our second DFAS review. Our first one was very successful. Our second one seems to be going just fine, and we continue to share with them our growth expectations over a 3-year time horizon, staying in close contact with them, and they have been very good partners.

The CFPB, as you know, did their first review of us last year. In their review of our operations and trying to evaluate the use of their scarce resources have chosen not to review us at all this year. And we're on notice that they will be back to see us not for the first 3 quarters of '18, but we'll see them in the fourth quarter of '18. That periodicity reflects their confidence in our servicing our customers.

In the market frame, where we think about our customers and competition, the evolution of the market, we have seen that our customer satisfaction continues to increase. Competition appears to be steady at this particular moment. And we look for our evolution.

One of the items that we've talked about a great deal with all of you folks is loan consolidation activity. And so we've been tracking it, and we have looked at it very carefully both by quarter as well as by competitor with whom our customers are interacting. I'm gratified to say that the third quarter loan consolidation activity is essentially flat, at a couple of million dollars, of the second quarter. So we believe we're at a plateau there, and that is something we will continue to monitor very closely, but it didn't grow as some people expected.

Our outlook. We're affirming that we will have $0.72 of earnings per share for the year, up from $0.53 in 2016, a 36% increase. By a happy coincidence, 2016, as I just said, was $0.53 a share, up from $0.39 a share the prior year, 2015. So we have back-to-back 36% increases in our EPS. [Grossly] compounding works in your favor. It gives us an 85% increase of our forecast 2017 EPS in regard to our 2015 EPS.

The outlook for originations. We're at $4.8 billion, a 3% increase, as I noted earlier. We, of course. We, of course, would have preferred to hit our $4.9 billion. It does look as though the market has grown slower than we thought. The market, as with all markets, fluctuates. We continue to improve our market share.

Having said that, we're not complacent about this. And as we follow our students, both in school as well as postgraduate, we also will concentrate more heavily as we go forward on the graduate funding.

And so for 2018, we will be introducing 6 new targeted graduate funding products. They will include business schools, legal, medical, dental, health professionals and a general graduate loan. And so as we look at this, we feel very good about one, our competitive stance; two, we've spoken to many, many schools about this, they are welcoming the additional option; and three, it represents the largest expansion in our market -- in our product set since before the spin. Additionally, as a validation of the fact that the ATLOS system that we put in 2 years ago, where we, at that time, told our investors that this would allow us to ease product introduction, this is clearly fiscal evidence of it, as we will be ready long before the 2018 busy season. So that's EPS and originations.

The third piece of outlook is the efficiency ratio. We are affirming that we will between 38% and 39% in the efficiency ratio. This, of course, is down from 2016's 40%, which is down from 2015's 47%, which is down from the run rate at the end of the fourth quarter after the spin of 51%. So continuous improvement is, of course, what we're interested in.

So the business pace continues. We continue to deliver solid results, with impressive growth and returns. We continue to improve our products, to expand our product set, to improve our services. We continue to have a financial profile and returns that are not only high, but improving.

The regulatory relationships are very good. Washington developments have been unclear, but as we look at those, one is they continue to be unclear; two is that we see possible opportunities. You all know about corporate taxes. We, of course, are a [full 3] payer. And if there is any expansion in the private business, we, of course, would benefit from that.

So we're still at the early stages of our story. We've gone through the launch, the establishment, the effective -- the efficiency -- effectiveness, efficiency, balance, and now we're looking for continuous improvement, providing consistency.

We continue to focus on 3 main vectors in regard to our investors: one, is attractive growth in earnings, where we are targeting the mid-teens; two, is operating efficiency, where we look to be consistently down, approaching as we've talked about with many of you individually, 35% or mid-30s in parallel, and to provide excellent returns on equity in the teens.

So thank you all for your attention. Thanks for your continued involvement with us to help us be a better player. And I'll turn this over to Steve.

Thank you, Ray. I think you've covered the landscape pretty thoroughly here so far this morning. I'll just add a few more details to the quarter's results.

Ray commented on the NIM and NIM being very strong for the first 9 months of the year, and we think that, that will continue, and NIM for the full year will be in the low 5.90s. So probably 10 basis points better than where we thought it would be at the start of the year.

A comment on the effective tax rate in the quarter. It came in at a low 34.7% compared to 45.5% in the year ago quarter. While the sharp decrease from the prior year was primarily due to changes in our provision for uncertain tax positions, I would point out that the artificially low rate in the third quarter did not contribute to EPS as it was an offset -- there was an offset running through the other operating income line of roughly $4.5 million.

Tax rate, as you've seen, will continue to move around quarter-to-quarter. But excluding these types of items, we think that a good run rate for the -- tax rate for the company is right around 39%. Other income totaled just $4 million in the quarter, compared with $11 million in the prior quarter and $22 million in the year ago quarter that had a number of onetimers. The $4 million was offset by the tax impact. So $10 million is a pretty good run rate for other income for the company on a go-forward basis.

We do like to provide investors with the default rates on loans that are just in full P&I. That came in at 2.11% in Q3, which was down from 2.24% in the prior quarter and 2.07% in the year ago quarter.

As Ray mentioned, credit is performing well within our expectations. Talk about the provision a bit here this morning. Provision came in at $53 million in the quarter compared to $41 million in the year ago quarter. The increase was primarily a result of an additional $2.5 billion of loans in repayment that we've added to the balance sheet. But there was an offset of $9 million that we mention in our 10-Q.

Basically, this adjustment included a number of both positive and negative changes to our forward-looking assumptions. They included a downward adjustment to our expected life-of-loan losses on TDRs, and offsetting that was a deterioration in the prices that we expect to receive on defaulted loan sales.

We ended the quarter with an allowance for loan losses of 1.33% of total loans, well within our trends and expectations. Allowance coverage ratio is a solid 1.9x charge-offs.

I think that, that pretty much covers the landscape. So let's open the call up for Q&A.

(Operator Instructions) And your first question comes from the line of Sanjay Sakhrani with KBW.

I've got a couple of questions. One is on the industry growth. I appreciate all the color, but I was wondering if you could just talk about what do you think that is driving that. And then on a related matter, if you were benefiting, as some of your competitors have retrenched for their own purposes? And then secondly, on the consolidation pace declining, that was a good sign. But as we're expecting potentially another rate increase later this year, do you expect that rate of consolidation to rise? And would you have any other or different competitive response if that were to be the case?

Do you want to cover the first one? I'll cover the consolidated one.

Sure. Yes, the industry growth, as I said, after the first 6 months of this year, as reported by Measure One, was 1.27%. It's driven by a series of factors. Of course, the demographics are the primary one. How many Americans are graduating from high school, and what is their relative propensity to attend college? And that number has been close to flat as far as graduates. The propensity to attend college has increased over the decade, but it's been relatively flat lately. Second is the availability of alternative financing that the family might have, remembering that over 90% of our loans are typically cosigned for first-time undergraduates. And also driven by the cost of the education. And so as has been widely reported, in addition to the demographic trends, the general balance sheets of Americans have improved quite a bit. And so their financing alternatives today are more attractive than they were 3 or 4 years ago. There's more equity in housing. There's higher balances for savings. And so in general, we are in competition with some of those alternative funding, whether it's out of savings or an alternative borrower. So that's -- so we combine that with college growth, which has -- they've tried to keep the inflation of education loans or education providing down, but they have not been 100% successful. And of course, there's normal fluctuation. And so as we look at this, it's a variety of items. They do have noise in them. We do believe that there will be continued growth in the core undergraduate market. We think that it will be somewhere between 3% and 4% when we add it all up, including our aspirations to gain some market share for ourselves. And so we'll look at that. The grad space has grown, of course, faster, and so our entry into that should provide us with an opportunity that is incremental to our trend line. In regard to the consolidation loans, Steve, I'll let you handle this, because the dynamic for the rate increase both affects the customer as well as the competitors.

Sure. Yes. So Sanjay, obviously, the consolidation loan business is very impacted by the level of interest rates in the marketplace. Consolidators are basically looking to offer a low rate to those who have primarily federal loans and then private loans as a secondary goal. To put that all into perspective, one year ago, the 3-year swap price, which was sort of the base for the funding that consolidators raise in the securitization market was at 110 basis points. A month ago, it was at 175 basis points and as we sit here today, it's at 191 basis points. So there has been a rise in rates, which obviously diminishes the returns on consolidation portfolios. We have not seen consolidators raise rates yet, but on this trend, it is going to be something that they're going to need to do to maintain profitability, and obviously, as they raise rates, that diminishes the size of the addressable market to consolidate. So it is worth watching. And I think the rising rates have already had an impact, and that might be why you see consolidations leveling off in our book today. I think your question was, will we see more consolidations, I guess, running up to additional interest rate hikes? For my money, I think the consolidation market is pretty saturated already. You can't turn on a sporting event without seeing one consolidator or another advertising. So it will be interesting to see. Clearly, we're seeing the trend somewhat muted in our book. But we have also talked about the fact that it is driven by fluctuating payment rates. We have another one coming in later this quarter, and we will see what happens with that. Regarding consolidation products, so clearly, the consolidation product as we know it, what we're seeing so far, as the world do, is a fairly low ROE business. So we are not interested in consolidating other federal loans, but we are very keen on protecting our portfolio. And we are looking to develop a product that extends term to reduce the borrower's cash payment. And we think at the end of the day, that's what really drives the demand for the product. We would hope to roll something along those lines out in the first quarter or second quarter of 2018. So we'll keep you posted on that front.

And your next question comes from the line of Shane -- Rick Shane of JPMorgan.

Look, there's an evolution that's going on here as the balance sheet builds, but the underlying trend is the shift in higher percentage of loans that are in repayment, from last year, you guys were at 64%. Now you're 67%. I'd like to just sort of explore, over the longer term, what you think the run rate for loans and repayment will be. How long it will take you to get there and then what the efficiency ratio, and you alluded to this, about a 35% efficiency ratio and what the reserve levels will look like as a percentage of loans when you get there.

Sure. So I have mapped out in front of me just the next year, but I think that's the trend we believe will be pretty consistent. So what we really focus on is loans in full P&I. That's where the real cost of service kicks in. We ended the third quarter at 32%. We'll end the fourth quarter of 2018 around 40%, and I'm going to guess that's going to creep up to maybe the mid-40s by the end of 2019. When we map out our full year financial plan, we think that the efficiency ratio should continue to trend lower and get into the mid-30s, all things being equal.

And what about reserve levels?

Oh, I'm sorry. The reserve levels. So look, we're at 1.33% today. I think on the last call, I talked about us being in the vicinity of 1.40% by year-end and then creeping up to the 1.50% level. We're not going to give you guidance today, but the trend is going to be for the reserve to creep higher as we have more and more loans in full P&I. So Ray talked about the fact that our run rate delinquency rate should be up to 2.3% now that we have a substantial portfolio of loans in P&I. Loans don't charge-off immediately. There is a loss emergence curve. So we need to continue to reserve the loans that go into full P&I. So I say I'm not going to give guidance, and then I take the bait. So are we going to be in the low 50s in 1.5 year from now? That's totally possible.

And your next question comes from the line of Arren Cyganovich with Citi.

Thinking about the grad school loan comments that Ray was discussing and targeting those loans, I just want to think about how that would work from a -- the risk of consolidation, because that seems to be the area that a lot of the loan consolidators are focusing. So what would you do from that perspective in terms of being able to competitively work with that rather than originate the loan and then have to lose it at a higher rate?

Sure. And the competition for a customer's balances is, of course, permeating from undergraduate through full P&I, as Steve said, and into graduate. We think the competitive frame is very similar. The federal program, First Grad PLUS, as you know has a 4.2% origination fee. We think that's off market. We think we can offer an attractive rate that -- where a customer will have a choice of either fixed or variable. We intend to match the term with the grad schools so far as what's normal for, let's say, medical versus business. And we expect to be fully competitive. I don't think that the frame in which we view the grad space is materially different from either the consolidated point of view or from the customer's point of view. And so as Steve said, the consolidators are on a narrow NIM, and as rates increase, that NIM will decrease because they're largely consolidating fixed loans. And so in offering the customers the choice and in trying to be competitive, we, of course, will have customers who are viewing us for the first time in the grad space. And to the extent we can make the offer attractive and they take that, I don't think that in any way increases the risk of consolidation from those balances.

Okay. And then just, I guess, staying kind of similar to that topic, Navient acquiring Earnest, can you address the risk to consolidation of Sallie Mae loans from that seemingly kind of stronger competitor under the Navient framework?

Sure. Two pieces there. One is Earnest has been up and running and been a competitor for as long as Earnest has been around. And so the idea that they are -- will be different, I think it may be in degree, but it won't be in kind. Secondarily, we have a noncompete with Navient, which has not changed at all since before the spin. It calls for, in a time period between now and 12/31 of '18, for them to be not a competitor in education loans, specifically a competitor for private student loans and to not consolidate our loans. That hasn't changed at all, and we always thought that Navient, once they were past the expiration of the noncompete, would attempt to be a competitor with us. By choosing the vehicle of going through Earnest versus doing it on their own doesn't particularly change what we think about a combination of the 2 of them as competitors, and we'll see.

And just lastly, did you ever consider bidding for Earnest or any of those? Or creating your own consolidation program?

So we certainly couldn't comment on the former, and we've talked, I think, quite a bit about our interest in a consolidation program. We are not interested in an interest rate-driven, low ROE business, but we do -- we are looking for ways to protect our precious loan balances and maintain them on our books. So I talked about that earlier, where we're looking to extend term to reduce monthly payment requirements, and we think that, that might very well be an effective way to compete around the edges with the consolidators.

And your next question comes from the line of Steve Moss with FBR.

With regard to the graduation -- graduate program you're rolling out, I heard you say it was incremental to originations. I was kind of wondering if you could put a little bit more numbers around that. Is that a couple of hundred million or several hundred million? How should we think about that?

Well, you should think about it as one, from a quadrant standpoint, it is a gross opportunity that turns into a net opportunity. Two, is we are attempting to forecast that now, and as I mentioned in the introductory remarks, we've had conversations with dozens of schools in regard to this. And so we are ball parking it. And I think as with most new products, it will start off relatively slowly and build over time. We are largely in a relationship business, both with the households as well as with the educational institutions. And so we're not at this point giving guidance on this, but we are certainly not looking to change our credit standards or change our general approach to the quality of our portfolio. And so we think it will start up at a reasonable level. And there's a big market. As I said, I think the government competitor profile has some weaknesses in it, and so we have high hopes, but we're not going to give quantitative guidance, but it will not start with a big bang.

Okay. That's helpful. And in terms of delinquencies here, I was kind of wondering -- it seems like given where -- it's trending higher, so I was wondering where do you think it peaks? Or how should we think about it over the next 12 months or so?

Sure. So regarding delinquencies, I just want to reiterate that we do think that the 2.6% was impacted by the hurricane situation down in the south. And Ray also alluded to an operational issue where we were not quite prepared desk-load-wise for the May/June repay wave, which was pretty sizable. And we are on a path right now, as we speak, to beating that back down into the 2.3% vicinity. We think that, that is the run rate for the amount of loans that we have in full principal and interest repayment now, and that's the number that you should hold us to and judge our performance by over the next several quarters.

(Operator Instructions) Your next question comes from the line of Mark DeVries with Barclays.

I heard your comments that you expect the efficiency ratio to gradually migrate down to the mid-30s. But as you roll out these new grad-focused loans, are there going to be some investments there that may cause a temporary spike up in the efficiency ratio?

So you know what, Mark, the interesting thing here is we've talked, I think, on prior calls that we were preparing for the potential elimination of Grad PLUS. So we have already made the investments necessary to roll this product out. And unfortunately, we did not see that elimination. But a positive byproduct of that is that we are prepared and ready to go without any additional major expenditures.

Okay. That's helpful. And then, Steve, I think you alluded to a couple of things that the deterioration in pricing on defaulted loans and changes in the adjustments and your expectations on TDR defaults. Is that kind of a more of a onetime adjustment that flowed through the provision this quarter? Or should we expect a certainly higher run rate? Is that more of a recurring item?

So Mark, it was actually positive. So if there are forward impacts, it will be to reduce our necessary provisioning. We're always tweaking our models and our outlooks, and we've talked in the past, I think, about the fact that we thought that our life-of-loan TDR loan loss estimate was probably a little lofty. It's based on primarily signature products. And one of the impacts this quarter is as we introduce more Smart Option data into our models, it does have a tendency to reduce loss expectations because of the simple fact that the Smart Option Student Loan product performs better. The pricing of defaulted loans was also another factor. It was less impactful than the TDR loan loss provision. So net-net, if anything, there would be a small positive to the bottom line in future quarters, but I wouldn't get excited and build too much into your models at this point in time.

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

Great. Actually most of my questions have actually already been asked and answered. Maybe just to kind of refine some of the answers. Ray, you had mentioned the noncompete with Navient, does that now extend to Earnest until 2019?

To the extent that Earnest is a wholly owned subsidiary of Navient, the noncompete will apply to that at the 100% level, yes.

So effectively, it will delay any competitive impact that they might have been having?

I think we can -- yes, to the extent that they are restrained by the agreement that they signed, yes.

Okay. That's interesting. Yes, and maybe you could just talk about -- when -- the comment you made about the grad market, just the fact that you're launching these products, I mean do you think that there's going to be -- is there potential for administrative relief even if there isn't regulatory relief on that? Can you talk a little bit about how that process might happen over the course of the next 6 to 12 months?

Well, we certainly are not planning on relief from either regulations or new laws. And so as we look at the graduate market, one is it's increasing; two is we have very small participation in it; three is that grad schools worry about fluctuations in the possible profile of Grad PLUS because they are dependent upon it for funding. So they're very interested in hearing about alternatives under scenarios which may be disadvantageous to them. So we are getting a welcome reception as we talk to the schools. But our thinking about this, our product pro formas, our ROE calculations are not dependent upon any change in the rules at all as of this moment. And as you allude, most people would say to the extent that there would be either rule changes or law changes, they probably will represent some sort of volume opportunity for us. And so as Steve said, in regard to thinking about that, we've done the investment for this year. We do have the extra capacity to handle this. But we're certainly not counting on it.

Right. Okay. And just last thing, maybe just an update on kind of on some of the nonstudent loan products and your thoughts kind of maybe going into 2018.

Sure. And so as I said, one is the increase in the grad space. I should just say that the introduction of the 6 different targeted grad profiles for products will be the largest increase in our product offering that we've had since before the spin. Secondarily, as you know, we are interested in the personal loan space because a student loan is essentially a highly tailored installment loan. We have set up the mechanics for doing a personal loan. And as Steve has said on prior opportunities, we expect to introduce that in a pilot form during the fourth quarter of this year so far as very limited testing, basically of systems, and then to have a pilot -- pilot offerings that will go out in the first quarter and then we'll evaluate both the cost to acquire from a marketing modeling standpoint as well as the distribution of risk with the idea of formulating underwriting targeting that would be appropriate for that market. So it all comes down to, yes, we will introduce a personal loan and we will do it very slowly through '18.

And your next question comes from the line of [Ann Mozeck] with (inaudible) Capital.

As you look at the overall performance of the company, the positive trajectory and all that you've delivered. And then you look at -- I'm looking at a metric in terms of your preferred dividend coverage of 26x. Does a single B rating make any sense to you? Because it certainly makes no sense to me.

Sure. Steve's spent a little bit of time thinking about this.

Ann, we think that the rating on the company is highly inappropriate and should be much higher based on the quality of the assets, the level of capital that we hold, the conservative funding approach that we take. But...

And of course, the track record.

And of course, the track record, which encompasses all of the above. The fact of the matter is we are held hostage by the rating agencies due to 2 factors that they put out there all the time. One is the monoline nature of the company, and two is the fact that they believe that retail Internet deposit markets is an untested and an unproven paradigm. I like to jokingly say that, that's like saying that Amazon.com is an unproven business model at this point in time. We continue to just stick to our knitting, improve the quality of the assets, and now we do have -- we have -- actually, I was reviewing it yesterday, we have 7 solid years of retail deposit gathering and we have a lot of CECLs on the decay in those deposits that's all very impressive, and now, of course, we are going through an interest rate cycle that proves that retail deposits are here to stay. And we've actually been growing that deposit rate throughout the last 9 months. So sooner or later, they're going to have to capitulate and agree that the we are an investment grade, if not better. So hopefully we'll see a couple of synergies with my friends at Moody's and S&P, but we'll continue to work closely with them to help them recognize the quality of the company.

And your next question comes from the line of Henry Coffey with Wedbush.

Could we revisit some of the loan quality comments you made? You went sort of through the walk through from 2.20% to 2.60%. Can you -- can we go over that again? And then I had a couple of questions around that.

Sure. Let me start, and then I'll turn it over to Steve. And so the 2.20%, let's start with that. And we said to you if we just changed the weighting of the P&I maturation pieces, and we're talking here in, obviously, small numbers of basis points, that we'd go from 2.2% to 2.3%. We also think that because of the substantial numbers of our customers who were affected by the difficult weather conditions in the South, but especially Texas and Florida, we think that, that has caused a transient increase of about 10 basis points. That would bring us to 2.4%. And so the actual was 2.6%. The seasonality associated with the maturation curve over this past summer caught us a little wrong-footed. And so we had lighter coverage in our collector workloads or desk loads, as Steve said earlier, than would have been ideal over the summer. We have corrected that, and we think that's about 20 basis points. But as we enter the post 3Q period, which we're 15 days into, or so, from a posting standpoint, we expected to see those delinquencies recede. And as Steve and I have both said, in fact, they have. And so we're on model. We were off model a little bit for weather and for seasonality adjustment.

No. So I was just going to add more operational details. So for people that are in these hurricane impacted areas, if they're in delinquency buckets, we don't advance them, so that keeps them in the 30-day and swells it. And that's where we saw the impact this quarter. And we respectfully did not call them. And we also, this time around, you might have noticed one of our -- the other student loan companies put people in these affected areas into forbearance. The last time we did that, we were met with the ire of our customers that said why'd you put us into forb; we don't want to be there. We're perfectly capable of making these payments. So what we did was we left them in the delinquency buckets, made no change to their status, and now that situation...

But we provided them with the opportunity, if they chose to.

Now that the situation is normalizing, we are contacting them. Payments are being made, and we are getting back on track.

And how quickly do we sort of get back to your target ratio of 2.30%?

We expect to be there in very short order. And if we're not there at the end of Q4, we will be having a different conversation. But we're very confident that what this is, is operational and not a change in overall credit. I mean, the other metrics, forb is in fine shape. Defaults are well within our expectations. While the seasonality was muted in full loans in the 1.08%, it is evident in the metric of 2.11% versus the 2.24% full P&I. So we are feeling good about the way this portfolio is performing.

The operational issue you rose about the summer activity, is that -- that's from loans that have been on the books for a couple of years that are jumping into their full payment mode? Or what is the dynamic there that kind of caught you?

So we had $800 million of loans go into full P&I in May and June. And on this call, I have given kudos to our collections department. They've been doing a very good job. Their cure rates are very outstanding. In this particular instance, the desk loads in the 30-day collection bucket rose. So in other words, they did not have enough feet on the ground. That has been corrected. We just graduated a class of 50 new collectors, and as we speak, they are busy working with customers in curing their delinquencies.

That -- and then looking on the growth side of the equation, the graduate school program kind of ramps in slowly. When is the first big year likely to be for that set of loan products?

Well, it's that number that we -- or it's a time that we hope would be, let's say, the year after this. And let's remember, though, that the graduate sort of dynamic is different from undergraduate. And so the graduate schools are smaller, are targeted and frequently run in a highly decentralized way, especially by major research universities. And so we will be picking up relationships with them and cultivating them through '18, and we're hoping that by '19, we will be full-fledged in regard to this and be a permanent alternative for graduate students in United States.

And then finally, going back to the credit rating issue, I guess I look at it from the point of view of dividends. I agree with the comments made. This is a much better company than a single B rating. I would also think, given your ROEs, I know you're still in a pretty aggressive loan cycle. When does that balance turn from loan growth and capital build to dividends? Or how soon? How about right now -- no, I get that you have a phenomenal balance sheet. What's the slowdown? But I thought I'd appreciate your view on it.

So look, all things being equal, on the trajectory that we're on, we do turn, make the inflection, from consuming capital to generating capital. The next wild card out there, which enables me to not answer the question forthrightly is CECL, which comes along in 2020. It is unclear what happens to the enormous loan loss allowance that we build to comply with CECL that currently would not be included in capital. So that is the next bridge that we are going to have to cross. But suffice it to say, the company does move from capital consumption to capital generation within the next couple of years.

But just appending that, in this case and all of our forecasts over multiple years into the future, it is the case that our ROEs remain very high, and it's our intention to continue to invest in the business as long as we can provide our shareholders with returns on equity that are respectable on the one hand and above available alternatives on the other hand. So that will drive us to continue to invest and not to consider dividends until such time as our ROE starts to diminish.

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

Most of the questions have been asked. I guess, one final question on credit. Steve, you talked about where coverage rates are going so far. I'm just wondering your guys perspective on what's a long-term, like, cyclical average loss rate for you guys when you have a stable loans and repayment position.

So I mean, the reality is when we look out over the next couple of years, there is so much [slots] in new loans coming on the balance sheet and more seasoned loans going into full principal and interest repayment. You get down to the conversation of what is the cumulative loss rate on this portfolio? And what is the loss emergence curve look like? And I'm going to say in big round numbers, cumulative loss rate in the 8% vicinity, and [50%] of those losses won't occur in the first 3 years of full principal and interest repayment. So there's no simple answer that 2.5% is the appropriate loss rate on this portfolio, because there's going to be so much change in the portfolio over the next 3 to 4 years. It sounds like I'm dodging the question, but I think it really comes down to the important factors here of what's the cumulative loss rate and what the loss emergence curve looks like.

But so that cumulative loss you said, that's what you would expect a normal cumulative loss factor to be then?

I mean, on the portfolio as it currently sits today, the cumulative loss rate left in the portfolio is probably in the vicinity of -- I'm thinking -- on the slide here, we've had a bunch of cohorts that have already experienced some defaults. It's going to be under 6%. So if we wanted to one-off now, which, obviously, we have no plans of doing, that 1.5% would probably be an appropriate loss rate.

Okay. That's very helpful. And then just a final question. Any thoughts on deposit betas? Where they've been? Where they're going to next year in your opinion?

So in our interest rate sensitivity disclosures, we use a beta of 0.85. Year-to-date, we have seen a beta of closer to 0.35, which has been, obviously, very pleasing. It seems hard to believe that in -- if there are additional rate hikes that we're going to experience such a low beta. I would say, my gut, it's going to be somewhere around 0.6. And if and when we have more data on that front, we would, obviously, change the model and give you detailed disclosures.

And your final question comes from the line of Michael Tarkan with Compass Point.

Just following up on that question. How do we think about the funding side of the equation at this point? Last year, you had done, I think, 3 securitizations. We've done one thus far this year by my count. So just kind of curious how you're thinking about that piece as we see sort of broker and retail deposits keep going up?

So Michael, very good question. When I look at my 3-year plan, I am looking 80/20, 80% deposits, 20% ABS. We do have a securitization in the works that will probably start to get marketed next week, and hopefully they get it up and down during this benign split environment. I'm sure they will. There's nothing on the horizon that's going to change that. We do like ABS because it enables us to extend the duration of our overall liability book. And we will continue to access that market, but we are a bank, and we will take advantage of the benefits that come along with that. I mean, the deposit markets are very deep and very efficient from a pricing standpoint. So that will be where we draw the balance or the majority of our funding from.

Okay. And then just the trajectory in the brokered deposits line. How should we think about that? I mean, is it -- in this kind of a consistent rate environment or if the fed raises in December, like how -- is that number going to keep going up at this pace?

So the ratio has basically been 50-50, brokered to retail and other. We had a big push on retail deposits this year, where we raised a bunch. What we target more is the ratio between brokered and retail, and we're very comfortable with that remaining right around 50%. If it goes to 45% or 55%, that's all good, too. But the 50-50 brokered/retail deposit mix is perfectly sound.

Okay. And then lastly, do you have what the new origination yields came in at on an average basis? I know the full book was 8.5%. I'm just wondering where new yields are coming in.

I think it came in right -- I do have it at, I'm going to say, right around, consistent with where the prior year's originations had been, as it's set to LIBOR, on the variable. And the fixed rate product is basically, given the steepening yield curve, or flattening yield curve, is basically right on where last year's originations came in. And we were well prepared and funded for that. So that has helped our -- helped us pick up a few basis points in them, which is a good thing.

And there are no further questions.

Thank you all very much for your attention on this call as well as the research that was done which prefaced your questions. We are very happy with the results of the third quarter. It's another solid quarter delivering good results for our investors as well as for our customers. I'm very much looking forward to the fourth quarter and 2018. And so with that, thanks again.

Great. Thanks, Ray. And 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 further questions, feel free to contact me directly. This concludes today's call.

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