SLM (SLM) Q2 2019 Earnings Call Transcript

SLM (SLM 2.86%)
Q2 2019 Earnings Call
Jul 25, 2019, 8:00 a.m. ET

Contents:

• Prepared Remarks
• Questions and Answers
• Call Participants

Prepared Remarks:

Operator

Good morning. My name is Adrian, and I will be your conference operator today. At this time, I would like to welcome everyone to the Sallie Mae 2019 second-quarter earnings call. [Operator instructions] Thank you.

I would now like to turn the call over to your host, Brian Cronin, vice president of investor relations. Please go ahead, sir.

Brian Cronin -- Vice President of Investor Relations

Thank you, Adrian, and good morning, and welcome to Sallie Mae's second-quarter 2019 earnings call. With me today is Ray Quinlan, our CEO; and Steven 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.

We will also refer to a new non-GAAP measures we call our adjusted earnings, which we plan to use as a primary non-GAAP performance metric upon the adoption of the new CECL standard in January of 2020. A description of core earnings and adjusted core earnings, a full reconciliation to GAAP measures and our GAAP results can be found in the Form 10-Q for the quarter ended June 30, 2019. This is posted along with the earnings press release on the Investor page at the salliemae.com. Thank you.

I'll now turn the call over to Ray.

Ray Quinlan -- Chief Executive Officer

OK. Thank you, Brian, and thank you all for your attention today. We have a bunch of stuff to talk about, and that will include a good earnings report for the quarter, as well as Brian has alluded, comments about CECL and adjusted core earnings. Turning to the quarter, it was another good quarter for Sallie Mae. Our volumes were up nicely at 9.2% faster than the market, we think about twice as fast as the core market actually.

The new credit quality is stable, as you can see in our numbers. Expense management and delivery of same continues to show efficiency both in core leverage of growth of revenue versus growth of expenses, as well as showing up in the efficiency ratio, which is another view of that. EPS growth continues to be strong, and our return over 19% again are very, very, very, I should say excellent. We continue to strengthen the franchise in addition to these results.

We are fully in the cloud as we speak. We are fully on agile as a regimen for getting things done, we're doing over 40,000 chats per month with a 91% customer satisfaction. We launched our credit cards during the quarter, when we renewed our brand with an effort that was initiated on May 6, which has been well received. And in addition to that, I'll make some comments about customer feedback from research that we have done, including customer view of both our product, as well as how they're doing in life, which I think are quite gratifying.

But just to return to the financial results for the quarter. Volumes $532 million of originated -- new originated private student loans were up 9.2%, were up 8.3% year-to-date. Our full-year forecast calls for us to be up 7.2%. As you all know, we're in the midst of our busy season, so every day is exciting for us, and that will continue up to our around August 20.

And so we get to third quarter, we will be able to give the results of the bulk of our volume for the year. As we go to the credit quality associated with that $532 million of new originations, our FICO scores for the improved accounts are rock-solid at 746 this year; also 746 last year. So they haven't moved $1 -- 1 FICO score point, I should say. Cosigner rate at 86% is also consistent. Our NIM at 5.88%, as Steve and I talked about in the last call, there will be some geography changes with NIM because we're increasing our liquidity on the balance sheet.

And Steve mentioned in the first-quarter call that that would happen throughout the year. In fact, our liquidity in the quarter has moved from approximately 5%, which had been the case in previous years, to about 13.5%, adding about $2 billion in liquidity. And so that of course depresses the NIM because it's against a larger set of balances, of which many are a wash. As we look to EPS impact, it should be minimal, we believe it'll be minimal for this year and also going forward.

And so at 13.5%, there were a bunch of questions last time we were together, are we close to having this done? The answer is yes. We should recall, though, not to be lost in the geography that I've mentioned, that our net interest income for the quarter at $397 million is up 16% from last year, and that does reflect the earnings capabilities of the company so far as generating revenue. As it turns into efficiency, the efficiency ratio at 34.9%, we regard as just terrific, down fully from 38.3% last year, so it is 3.4% in the efficiency points down year on year, which is a steady improvement. And I will say that we're in our sixth year since the spin and this has been a story that has been 100% consistent through these years and very gratifying, moving from originally about 51% down to this 34.9%. Our credit performance. In the first quarter, our loss rate was 89 basis points.

I commented at that time that that performance was in excess of, that is exceeded our expectations in the sense of being low. As you all know, we typically have written off per quarter between 1% and 1.5% of our ongoing portfolio and that's adjusted both by seasonality, as well as how vintages come in and go out and weighted average changes in the life of the portfolio. And so when we see the 89 basis points, which is down 129 basis points in this particular quarter, and we talked in the first quarter that the 89 basis points was lower than our anticipation, 129 basis points is more in line with what we're doing. And when we see the year-to-date write-offs of 109 basis point this year versus 108 basis points last year at the same time, obviously, the two numbers are right on top of each other.

And all of these numbers fit within our general expectations as we model the portfolio. The delinquency, which has gone to 2.7% is also within our range of model, and that is up from 2.5% in the first quarter, fully, OK, the vintages as they mature through the season, we're watching each of those carefully. We see no real change in any of the performance. The PSL reserve, the private student loan reserve on our balance sheet, which reflects our anticipation of what we think the write-offs in the portfolio are is a good indicator of sort of our confidence in the model. And if we look at the 2Q '19, PSL reserve is 1.42%, 142 basis points, last year at this time, it was 140 basis points.

And so we are on model and the idea that the losses went up in the quarter and somehow that was unexpected, is not really what's was going on. The 89 basis points was low, the 129 basis points is still within the range. And so we are on model, not deteriorating any of our forecast and the portfolio is still in a maturation rate. So the weighted average movements have to be taken individually in viewing the model, and we are happy with the way those things are going. The balance sheet overall is growing at 22%, continues to be strong and continues to reflect the fact that we are buying -- we are originating, holding and servicing all of our loans.

EPS, which was just $0.306 in the quarter, up from $0.25 a year ago, 22% increase. Of course we like that, and we will continue that as far into the future as we can. And our ROE at 19.8% consistent with last year's 19.4% shows the high quality of both our assets as we originate them, as well as the efficiency with which we service them. Revenue cost growth reflects that as well, including just a graphic point of leverage for the franchise.

Revenue is up over the year before 15.3%, expenses are up just 4%. So we have a 3.5x ratio of expense growth versus -- of revenue growth versus expense growth. Our outlook. The originations at $5.7 billion were holding. As I said, we were at the early stages of our busy season.

We'll know at the end of the third quarter much more accurately what expectations are for the year. Over 9% going into the busy season is of course helpful. But busy season's filled with competitors, we'll see how it goes. The efficiency ratio guideline for 35% to 36% is consistent. EPS.

EPS guidance has been dropped. And so let me talk about that for a minute. There's a situation with our total debt restructuring accounting that runs into problems when the interest rates in the environment, especially the forwards, drop. And so this is the first time that we've had an interest rate declination of any magnitude since the company was launched in 2014.

When a total debt restructuring item dollar goes into our reserves, what happens is the LIBOR on the day it goes is frozen at whatever it happens to be. And then we take a look at that, we look at the forecast for the cash flow associated with that item into the future at whatever the APR for the company or for that particular account happens to be. So the APR is whatever it's doing, whether it's fixed or variable. But that piece of LIBOR is frozen from the day it goes into the reserve.

And so when we have an environment where interest rates are dropping, we have a compression on the future cash flows associated with total debt restructuring within our reserves. And I know this is a little bit arcane, but nonetheless has impact on us. And it's new because interest rates have not been going down before. We knew this forecast was going to occur.

We had in our original expectations some thoughts about that. But the forwards have dropped faster than we thought at the end of the first quarter. The result of that is that there is an after-tax $15 million negative hit associated with the TDR evaluations of the future cash flows, which is driven by accounting, not by cash flow in real life. And so that $15 million with our 430 million shares outstanding is a $0.035 negative impact to us this particular quarter for a calculation that as we get to CECL will disappear, and so it will only be relevant for the remainder of this year.

So it's in oddball one-off impact for us, but it is $0.035. So if we had our original guidance of the $1.23 to $1.26 and we lost $0.035, we would be down to $1.195 to $1.225, in fact, the portfolio and the business is performing better than we thought. And so we actually -- had we not had this TDR accounting sort of impact, we would have increased our guidance and all likelihood, as we have done in prior years, tightening up the guidance at the first half of the year that would've been up by $0.015 or so on the low end, about $0.01 on the high end. But in fact, that improvement was used to partially offset the $0.035.

So $1.23 to $1.26 was down by couple of pennies to $1.21 and $1.23. And so this is no real impact on the business going forward, no impact on the franchise. It is not a reflection of any deterioration in credit losses. It is an odd accounting regimen that is only extant for the next six months and only relevant to us as interest rates change and significantly down.

So I'm sorry for the wandering off to that, but that is an important piece of understanding that there's a negative impact, but it's not a franchise impact. So in summary, we have a credit card launch in of the quarter, which is very good. The new branding is done. The personal loan is now on track to have an ROE of 15%. Our 9% growth is faster than the market at 3% to 4%.

I should say also that the customers, as I alluded to earlier, value the products quite a bit. We've done research on private student loans, and just note some highlights of it. 91% of the private student loan borrowers have completed their program, over 90% are employed. 79% agree that borrowing gave them a better education than they would have had otherwise.

83% say their education has contributed to their career success as they sit and also over 80% are satisfied with their jobs. 77% feel successful where they are in life. And as a note, which I think is little bit unexpected in the political regimen in which we find ourselves, 43% of private student loan borrowers also had a Pell Grant. The idea that private student loans are somehow a class of people that are way above normal Americans is in fact, not true.

We have also done some research on how America pays for college. And 90% of customers -- or 90% of students and their families view college as an investment. Surprisingly 71% -- given all the publicity that's out there, 71% believe that the price of college is fair. 79%, by the way, under the heading of price shopping have eliminated at least one school as they searched for where to place their students and they eliminated at least one school because the school was too expensive.

And so we have a bunch of research about the current state of affairs in higher education, which is an industry-leading research piece, but has some very surprising results, which we will be sharing with politicians as we go forward. So we continue to strengthen our franchise, college is a great investment, the outcomes are very good. And I should note also that 9% that we mentioned as far as volume is increasingly segmented from the original charter that we had of undergraduate private loan funding. And now with our segments of six new products in graduate, parent, partner, the career training, just entered international, about 20% of our volume comes from what is nontraditional. Remain No.

1 in the market with excellent returns, controlled expenses, rational capital allocation, including $60 million of buyback in the quarter, return of capital is with us and leverage is important. Two items to cover now, which I will touch on and I know Steve will cover. CECL, and so CECL has two pieces to it for us. One is the initial impact, which Steve will talk about and which is consistent with our prior disclosure; and the second is CECL has an ongoing EPS distortion in such a way that, as I mentioned, we're in the midst of our busy season now.

Under CECL a year from today, if we have a widely successful third quarter, we'll be forced to fund the life-of-loan losses in CECL, under CECL for the quarter. And so perversely the better you do in sales, the worse your EPS is liable to be. Therefore, we've chosen to try and take what we think is a more representative core adjustment, which will be our GAAP earnings minus any impact to the loan loss provisions in that particular period plus the current losses experienced in that period plus the tax effects associated with those movements in order to come to core earnings, which in some sense are traditional EPS earnings. We'll do the -- we will forecast these -- we'll use these for forecasting guidance, we will disclose these, but we think it is over a period of time much more representative of the quality of the franchise than the CECL distortion.

So with those comments, I will turn the microphone over to Steve.

Steven McGarry -- Chief Financial Officer

Sure, OK. Thanks, Ray. So yes, as Ray mentioned, we added a significant amount of disclosure in our 10-Q this year as we prepare for CECL. CECL is all but upon us and we will be adopting it in essentially five short months.

So if you look at the critical accounting policy section of the MDA -- MD&A, we reported an estimated range of the impact of our loan loss allowance for both our student loan and personal loan portfolio as if we adopted CECL on June 30 of 2019. As we explained in the document, we publish a range because our loan loss models have not yet improved validation. They certainly will be by the time that we have to adopt the standard. However, the midpoint of the range is the estimated impact from the CECL accounting standard. And it's really actually not much different than what we've been sharing with investors from November of last year, I think we first put this in our disclosures.

But to put a finer point on it, across our student loan portfolio, we estimate that the billable life of loan allowance, we would need to add $1 billion to our reserve today, bringing the total reserve for the student loan portfolio to a lofty $1.3 billion or 6% of our ending total loan balance. For the personal loan portfolio, if we had to build a life of loan loss allowance today, we'd add $75 million to the reserve, bringing it to roughly $150 million or 13% of our total ending loan balance. So to build this reserve on day 1 of CECL adoption, the company will run the tax affected reserve build through GAAP equity, reducing it by approximately $800 million and that creates a deferred tax asset group tax impact of roughly $250 million. For purposes of calculating regulatory capital, banks are given the option of phasing in the impact on equity over a three-year period. We certainly will avail ourselves of this phase-in option. In addition, this is slight offset, banks can reduce risk-weighted assets by the amount of the loan loss allowance that doesn't qualify as Tier 2 capital, which slightly offsets the impact for us.

A significant amount of our loan loss allowance will not be included in Tier 2 capital. The important thing here is we will continue to be well-capitalized after the adoption of CECL following the adoption of CECL, we believe our true capital position is going to equal our GAAP equity plus our loan loss reserve. This is a ratio that will be well into the mid-teens and will actually grow during the three-year phase-in for the regulatory capital on our balance sheet. A couple of other points. To adopt CECL, we've built an economic model to forecast expected credit losses.

We have to adopt economic forecasts and adopt a baseline, reasonable and supportable forecast period. What we're going to do is employ Moody's economic forecast. We'll use a baseline, an economic downturn and economic upturn forecast and blend those for our CECL model. And what we're going to do is to use two years of forecasts and then revert to our mean expected losses for the remainder of the life of loans.

Adopting CECL is a very complex project, and we've been working on it for several years now. And we will certainly be ready when we need to on January 1, 2020. And as Ray pointed out, once CECL was adopted, there is going to be significant volatility in our earnings. So I think the pro forma number that we've crafted and disclosed an awful lot about in this quarter's Q is going to really get right to the heart of the company's performance on a go-forward basis.

So that's really a summary of what we disclosed in the 10-Q on CECL this quarter. So with that being said, I think it's time to turn the call over for Q&A.

Questions & Answers:

Operator

[Operator instructions] The first question comes from the line of Sanjay Sakhrani from KBW.

Sanjay Sakhrani -- KBW -- Analyst

Thanks. Good morning, and thank you for the color on CECL, Ray and Steve. I guess just some clarifying questions on that. So when we think about this impact that you have with the TDRs and into the run rate EPS ex your core number, is it fair to add back the $0.035 to your guidance range as a starting point? And then when we think about the impact of CECL to that number, not the non-GAAP number, what exactly is it?

Ray Quinlan -- Chief Executive Officer

I'm sorry, what was your second part of that question, Sanjay? We didn't quite follow you.

Sanjay Sakhrani -- KBW -- Analyst

Yes. What's the CECL impact to the actual number, to the GAAP number?

Steven McGarry -- Chief Financial Officer

So there's no CECL impact, obviously, for this quarter. But so that when we adopt CECL, the catch up reserve will be run through equity. So there will not be a hit to the income statement. But on a go-forward basis, for example, this quarter, when we originate whatever we're going to originate, $2.7 billion, actually we will originate $5 billion of loans, big round numbers this quarter because there will be the second disbursement.

We will have to hold a life-of-loan loss reserve for that entire origination tranche, which let's not get into the math of discounted cash flow model that we're going to employ, but that's going to end up with, call it a 6% reserve that's going to go through the income statement, which is $300 million, which is sizable. So that is why we think it is very important to introduce this core earnings impact. And I see that Mr. Quinlan wants to add a further comment here.

Ray Quinlan -- Chief Executive Officer

I think part of your question is given the increase in the loan loss reserves this quarter associated with TDR dynamics that I mentioned, whether or not that is impact on EPS going forward, and in particular as we adjust the CECL, how is that to be treated. And the answer I think is quite straightforward, that CECL will have its own adjustments for the life-of-loan and the associated forecast for that. And the increase in reserves associated with TDR, the $0.035 is in some sense a prefunding of CECL. So when we look at CECL to the extent that our loan loss reserve is higher this quarter than we had anticipated because of that LIBOR idiosyncrasy is sent into it, we will reserve less when we do this CECL change from our current reserve to the CECL change.

So when Steve mentions it's a $1.3 billion overall, within that there is an accommodation that says, oh, because you added the $0.035 this quarter you don't have to add it on January 1, 2020, and so it's part of the mix, Sanjay.

Sanjay Sakhrani -- KBW -- Analyst

Right, right. And that's helpful. And then as we think about future rate reductions, to the extent there are any, hopefully not, does that mean that the TDR impact goes away after that? And all we're looking at is provisions related to originations?

Steven McGarry -- Chief Financial Officer

So we've got a significant amount of interest rate reductions in that TDR number today, I think it's 62.5 basis points, to the extent that rates declined more than that between now and end of the year, there would be an additional impact. Although it's going to be small. But and then of course, once we adopt CECL, the TDR impact goes away.

Sanjay Sakhrani -- KBW -- Analyst

OK. Great. And I guess just one last question on that...

Ray Quinlan -- Chief Executive Officer

The benefit is we wish the Federal Reserve waited six more months before they dropped interest rates.

Sanjay Sakhrani -- KBW -- Analyst

Yes, I know. I hear you. And then a final question on credit quality, that obviously is really strong for some time now. Could you just talk about sort of what's driving that? Is it just the healthy economy which is driving that? Thanks.

Ray Quinlan -- Chief Executive Officer

OK. Well, one is thank you very much for framing the question that way because the earnings -- the write-off rate went from 89 basis points in the first quarter to 129 basis points. And I agree that the 108 basis points which is the year-to-date number for write-offs is an excellent number. We haven't seen much change due to the economic environment associated with our roll rates and other items.

We think we're still driven by the quality of our underwriting. And over the last three years, the most significant item that has been adjusted as we've done our forecasts for both delivery -- delinquency; straights, that is people who wind up not paying at all; and ultimate write-offs has been our ability to help our customers adjust from the graduation period to the full payment period. And so the classic undergraduate graduates in May, has to make a full payment for the first time in November. Now 50% of our customers are making either interest-only payments or $25, but that also will go up, of course, when they go into full P&I.

And we have been able to give them a better understanding of the calendar associated with that, what the dynamic will be, asking them if they're ready to pay, if they haven't gotten a job yet, working through any sort of a payment arrangement that would temporarily help them. But that particular piece of better customer communications and more aggressive upfront contact with our customers as the vintage matures has been the most outstanding item. The environment has been helpful, but the group that we're dealing with, remember, has an unemployment rate that is half of what the national rate is. So the unemployment rate for college graduates is running under 2% and hasn't been over 3% during this entire period.

So that particular demographic has been in a very healthy economic environment, but that's the same as saying that our underwriting is high-quality. So we think it's quite steady, to answer the question a long way around.

Sanjay Sakhrani -- KBW -- Analyst

Thank you.

Operator

The next question comes from the line of Michael Kaye with Wells Fargo.

Michael Kaye -- Wells Fargo Securities -- Analyst

Good morning. I was hoping you'd give some updated thoughts on your NIM expectations, just given the change in the rate environment. I thought last update, last quarter was about 5.9% for the year.

Steven McGarry -- Chief Financial Officer

Yes. Michael, we added liquidity a little bit quicker than we expected and we'll probably go a little bit higher than we originally thought we would. Ray mentioned that we're going to hover on that 13% to 14% vicinity. That will result in a NIM for the full year of roughly 5.8% as opposed to the 5.9% that we mentioned on the last call.

Ray Quinlan -- Chief Executive Officer

But interest income as an actual number will be relatively unaffected.

Michael Kaye -- Wells Fargo Securities -- Analyst

OK. OK. That's helpful. And then I had a quick question on peak season.

I know it's still kind of early, but from what you gather so far, I was hoping you could comment on a more broader pricing environment, particularly with some like newer upstarts like Navient, SoFi, guys like College Avenue, have you seen more pressure on pricing so far?

Ray Quinlan -- Chief Executive Officer

The pricing, there's two pieces to set a minimum, right? One is in general. And in general, the pricing has been quite stable and as mentioned in prior calls, new entrants who are rational, such as SoFi in this particular environment, have a pricing spread. And of course, there's a spread on variable, as well as fixed. Have a pricing spread that is on top of ours, very similar.

We have seen, as we get into busy season, a couple of people will lower the low end of their ranges so that they can have a 399 and narrower number like that. That's been relatively rare. The people who are newer tend to do it little more, and so this is a normal sort of trench warfare associated with the busy season. I wouldn't say it's unusual for us versus prior-year result.

Michael Kaye -- Wells Fargo Securities -- Analyst

OK. OK. That's helpful. Thank you.

Operator

The next question comes from the line of Moshe Orenbuch with Credit Suisse.

Moshe Orenbuch -- Credit Suisse -- Analyst

Great. Thanks. I did see that loan consolidations were down slightly from Q1. Could you just talk about whether that's a result of seasonality or a result of your competitors or things that you're doing? And what it might look like as the year progresses?

Ray Quinlan -- Chief Executive Officer

Sure. And the sort of the change in the, I think consolidation, it went from like 393 to 314 and now it was gratifyingly down. And when we look at the individual companies that make that up, every single company was down from the first quarter to the second quarter. Very hard for us to gauge what the intent is of companies who are doing this because primarily what they're doing is consolidating federal loans.

And as you know, private student loans only run about 10% of the volume of the federal loans. And so we've always been the tail on that dog. And we do think that several of the people who are in that who are not traditional banks in the sense of having a low cost of funds aren't making money at this. And we think that they are coming back to a point of view where their P&Ls will dictate that this is not a particularly attractive business.

But two things, one is it's nice to see it down; two is going forward, we're watching, of course, every day carefully. But we don't know what their individual motivations are. Clearly, some in the first quarter were just looking for volume. But that as you know is a short-lived type of strategy.

Moshe Orenbuch -- Credit Suisse -- Analyst

Got it. And my follow-up question, Steve, you had mentioned that you think that GAAP capital plus some form of the loan loss reserve is going to be the operating capital metric. Just can you talk a little bit about I mean have you talked to their -- I mean to the regulators? I mean, how do you think about that from kind of a regulatory standpoint? And other things that you are -- could do or are going to do if originations tended to be higher, I guess, than your expectations?

Steven McGarry -- Chief Financial Officer

So yes, look, we always discuss our business plans with the regulatory agencies. And when we do that, we show them longer-term capital forecasts and discuss things like share buybacks. And obviously going into CECL, both of those were on the table, and we certainly are returning capital to shareholders at this point in time. And obviously, we pay a lot of attention to our longer-term capital forecast.

And we do have a significant excess capital cushion. In the wonderful scenario where our originations increased 20% or 30% more than we're expecting, I don't think that that would put us into a position where we are eating into our capital cushion. So we're in fine shape given the outlook for the next couple of years here.

Ray Quinlan -- Chief Executive Officer

And if I could just add an addenda to that. It's the case that the regulators have not changed their guidelines for well-capitalized and other benchmarks that they have. It's also the case that everybody recognizes that this $1.3 billion or $1 billion that's moving from the equity account to the loan loss reserve account is really a cushion, as Steve says. So when you look at total loss absorbing capability or TLAC for the institution, it's appropriate.

And if we were buying the institution, we'd say, gee, what's the cushion against these assets not being high quality, you would add the loan loss reserve to the equity and you would that in the first 15 minutes of evaluating a company. So the regulators get that, they haven't changed their guidelines, no one has lived with CECL yet. We certainly are communicating to them that when we add the two numbers together, it's a number, as Steve said, mid-teens, call it 15% capital, extremely well-capitalized. And we have been straightforward with the regulators that we think that is a, one might say an extremely healthy number on the one hand; the other might -- you might say gee, it's much more than what was viewed to be an adequate cushion in 2018, let's say.

And so we haven't lived in the world yet, but as Steve says, under any circumstances, we're well capitalized, and we will continue to view capital allocation in a rational way.

Moshe Orenbuch -- Credit Suisse -- Analyst

Got it. Thanks. And I do agree with the idea of the cushion and capital.

Ray Quinlan -- Chief Executive Officer

All right. If you want, I can give the address of the FDIC so you can send in your vote of confidence in this.

Moshe Orenbuch -- Credit Suisse -- Analyst

Yeah, FDIC has some rules about that.

Operator

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

Arren Cyganovich -- Citi -- Analyst

Thanks. Yes, I just was curious, realizing that NCOs are still at a low level, but what's your expected trajectory going forward?

Steven McGarry -- Chief Financial Officer

Arren, we're having a hard time hearing you on the cellphone connection isn't very good, could you please repeat that question?

Arren Cyganovich -- Citi -- Analyst

It's actually my work phone. But it's -- the question was what's the expectation for the trajectory of your net charge-offs going forward?

Steven McGarry -- Chief Financial Officer

So typically, the second quarter is the peak charge-off quarter as new loans go into full principal and interest repayment in November, December. Typically, new borrowers struggle, and there's a chunk of zero payment defaults in the second quarter and then things tend to level off. So if year-to-date defaults are similar to last year, we don't think that '19 should be a whole lot different than '18 by the end of the year.

Arren Cyganovich -- Citi -- Analyst

OK. And then on the credit card side, I know it's still very early days there, can you just share any of the kind of feedback you've had since the initial rollout of that?

Ray Quinlan -- Chief Executive Officer

It is early days and we are still doing a very controlled solicitation. So the number of customers that we have while we're sitting here measured in the hundreds, right, just to put a point on it. We're very happy with the way the launch went. We had recently signed a -- where we signed our contract with our partners, Deserve on July 6, 2018, and we launched the product in June of this year, getting it just under the full 12 months.

And so we're very happy with that. We have three embodiments for the life stages that are in the product, the reviews by people who do these things has been positive. And so we're on the map and as you point out, early days, more to come in future quarters.

Arren Cyganovich -- Citi -- Analyst

OK. Thank you.

Operator

[Operator instructions] The next question comes from the line of Rick Shane with JP Morgan.

Rick Shane -- J.P. Morgan -- Analyst

Hey, guys, thanks for taking my questions this morning. Look, I understand in terms of charge-offs there's high degree of seasonality, and there's always a lot of noise. I am curious when we sort of look at some of the underlying drivers, is there anything you're seeing in terms of roll rates that has changed either positive or negative over the last 12 months?

Ray Quinlan -- Chief Executive Officer

As we look at each one of the vintages coming in, we have various collection strategies that we -- vary over time. There are always fluctuations, and so the questions would be as we touched upon a little bit earlier, is there anything from the environment, are the new accounts performing the way we thought the new accounts would be? How are first payment defaults going? How are contact rates doing? And the answer on all those things is everything is stable. And so as Steve alluded to, the write-off rate last year at this time was approximately 108 basis points. We're right on top of that 108 basis points year.

There's always puts and takes in collections. And as we commented in the first quarter, the 89 basis points that we wrote off was better than our expectations. The 129 basis points that we're doing this quarter is within our expectation. So I would say a bunch of noise around, but that is BAU for collections.

And no change in our models, no change in our cutoffs, no change in our through the door population. Stability has been our companion and friend for several years, and we expect the case to be going forward.

Rick Shane -- J.P. Morgan -- Analyst

Got it. And you made the comment that the new vintages that are rolling through are performing consistently with historic vintages. I'm curious if you're seeing anything in the more seasoned vintages that changes your trajectory perspective or your trajectory outlook there? And is performance -- you guys are moving into a world where you have to look at lifetime performance. Is there anything that you've seen over the last couple quarters on your more seasoned vintages that changes your outlook either positively or negatively? I remember a couple years ago, we talked about sort of 8% lifetime loss rates and then that actually got dampen down a little bit.

Curious if you are seeing anything as the vintages mature that changes as well.

Ray Quinlan -- Chief Executive Officer

Yes. Lifetime losses are something that we have built into our evaluation of our front-end and underwriting and cutoff scores. And so from the very beginning of this company's life, we have used lifetime losses. Now CECL will cause us to have to document those and put them on the balance sheet and that sort of thing.

But essentially that's taking preexisting models and putting them into a regulatory regime that is more about model risk management, that sort of thing. The models themselves have had very little change in them. And you would see it in our through the door cutoffs either in the approval rate, which is running about 45% or with that 746 FICO. We have not seen any changes, and I will say that to the extent that the consolidation numbers go down, it's very helpful for looking at lifetime losses because typically those are seasoned a year or two.

And so once again, the answer has been steady. Steady has been both our companion and our friend. We are dealing with a demographic that is not as volatile as the economy at large with the unemployment rate running half of what the national rate is, as I said earlier. And so we -- direct answer to your question, we haven't seen anything either in the new vintages coming in nor in the vintages that are past their peak losses, and as you know peak losses occur about two years -- or within the first two years of full principal on interest.

That tail off curve of losses is on our models, and so that's, as I said, had it not been you would see us adjusting the front-end.

Rick Shane -- J.P. Morgan -- Analyst

OK. Great. Thank you very much.

Operator

The next question comes from the line of Vincent Caintic with Stephens.

Vincent Caintic -- Stephens Inc. -- Analyst

Thanks. Good morning, and thanks for the disclosure on CECL. I think you guys have been kind of the thought leader on providing framework for CECL. So my first question is actually on it.

As we're looking at the 2020 when we have to look at a CECL EPS number and appreciate you'll be putting a core number or adjusted core number. But any sense for how the CECL affects the GAAP number? Is there maybe any -- because I understand you're growing a lot. Just kind of when we think about 2020, if there's any help on what the GAAP versus the core spread would be for EPS.

Ray Quinlan -- Chief Executive Officer

Go ahead.

Steven McGarry -- Chief Financial Officer

The impact would be bigger than a bread box, how about that. I mean the impact is going to be significant. We just talked about a $300 million reserve on an origination cohort of $5-plus billion. So it's going to be a meaningful number.

Ray Quinlan -- Chief Executive Officer

But we haven't given any guidance so far as we think that is. Although we do, as Steve said, clearly have in the camp of not only will it be highly volatile number, but we don't believe it will be representative of the performance of the franchise. It's extremely unfortunate that that is the case. But if you wanted to ballpark it, you can look at the originations in the quarter.

You could think about those life of loan losses in that high 7% to 8%, you would ballpark what's going into CECL versus what had previously been the case. And you would take it write-off our adjusted core earnings, and look at what we're backing out for the loan loss reserve build because that is the definition of the, in my opinion, inappropriate volatility that's a side effect of CECL. So we're not giving guidance on it, but it's an easily calculatable range given the disclosure that we've already provided.

Vincent Caintic -- Stephens Inc. -- Analyst

Got you. That's helpful. And maybe asking it another way. So you've -- your 2019 guidance of $1.21 to $1.23, I guess under the new adjusted core EPS, that would still -- would that 2019 guidance have been approximately the same, so there's not really any change going forward?

Ray Quinlan -- Chief Executive Officer

Well, we're not changing the guidance for any impact on CECL for the remainder of this year. And so is your question had we been under CECL in 2019, what our EPS would have been?

Vincent Caintic -- Stephens Inc. -- Analyst

Yes. Under your new adjusted core EPS method, just so we can have apples-to-apples for what you'll be discussing as your core number for next year.

Steven McGarry -- Chief Financial Officer

Yes. So look, if you take a look in the 10-Q, we laid out 2018 and '19 under current core and future core and the difference ranges between $0.03 and, I think $0.06 per quarter. So the impact is not significant, but once we compare that to a CECL number, the impact will certainly be significant. But all we're doing in the new core is rather -- when we give guidance, we will give guidance for the new core and instead of including provision, it will include expected charge-offs during the course of the year, which obviously all things being equal, are going to be lower than the provision because the provision is forward-looking and includes growth in the portfolio.

So you can take a look at the numbers and if you want to discuss it further, we can get together with you after the call. We did -- we published an awful lot on what the new core impact is going to be.

Vincent Caintic -- Stephens Inc. -- Analyst

OK. Got it. That's helpful. Just a follow-up separately.

On the personal loans, just noticed the credit there, losses jumped, and I was just wondering if there's anything particularly related to the quarter and what you -- what we should be expecting as a run rate on that loss rate going forward? Thank you.

Ray Quinlan -- Chief Executive Officer

Sure. And as you know, the losses that we are experiencing and the provision change that you've noted, one, starts at zero on November 1, 2017, and we built up the portfolio through '18 with a range of testing, both on the marketing side, as well as on the credit side. And so that's reflected in the provision. We've made adjustments appropriate to both marketing, as well as credit starting in the middle of last year.

And so going forward, we will have a provision that sort of models the experience of the personal loans in the current period, but we're not giving guidance for that individually. But it is the case that our portfolio going forward does have an expected ROE of 15%.

Vincent Caintic -- Stephens Inc. -- Analyst

OK. Got you. Thanks very much.

Operator

The next question comes from the line of Mark DeVries with Barclays.

Mark DeVries -- Barclays -- Analyst

Yes. Thanks. So given your comfort level with your capital cushion, even kind of post full CECL implementation, can you just talk about the capacity you think you have to continue to buy back your stock here, which presumably you view is pretty attractive at these levels?

Ray Quinlan -- Chief Executive Officer

One is I agree that the stock is attractively priced for buyback. I wish it was just attractively priced. But having said that, we're not giving any guidance on the other side of January 1, 2020 for CECL. Because there's just so many moving parts that have not been in any way blessed by regulatory bodies.

So we don't want to get ahead of ourselves. But as Steve says, going forward, to the extent we had loan loss reserve plus the capital piece and that is 15% or so going forward, and we are well-capitalized. We will continue to view the capital as a series of capabilities that can either be used to reinvest in the business, pay in dividends, do buybacks, and we would like to continue to do all three going forward. And so that would be our objective, but we do think that it's appropriate to let the dust settle a little bit before we would get ahead of ourselves in these projections.

Mark DeVries -- Barclays -- Analyst

Got it. And then on the refi, or the refi-related repayments. It sounds like you attributed a lot of the drop to competition kind of pulling back. Should we assume that you haven't had to do much as far as some of the defensive measures you contemplated? And how should we think about how that activity over the back half of the year with the rates dropping and the refis presumably looking a little bit more attractive for borrowers?

Ray Quinlan -- Chief Executive Officer

Well, first, it is the case that of course, if the this level drops and we attribute it to individual competitors when we have their names, they will drop in total, so that 393 to 314 drop. The thing about it that was interesting was a drop for every single name on our list. And so we don't know what they're going to do going forward, and as I said, it's been a thin to negative margin business by any calculations in the past. We haven't had to do much in the way of defense, to take another part of your question.

And so we haven't had to deteriorate in any our either yield on the portfolio or our net interest margin as we discussed here. Going forward, very hard to say. The drop off is significant, and we have seen over the course of three years that some competitors get in, they stay a while and they leave. And so we'll see what happens with them.

But it's entirely dependent upon their individual initiatives. But it is gratifying to see that across the board, they all seemed to think it was a better idea to do less of this in the second quarter than the first quarter.

Mark DeVries -- Barclays -- Analyst

OK. Got it. Thank you.

Operator

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

Henry Coffey -- Wedbush Securities -- Analyst

Yes, good morning, and thanks for taking my questions. Just three quick questions. No. 1, in the ancient days, bank analysts called what you're talking about primary capital and even though it's a bit of a misnomer, cash EPS. You obviously talk to your competitors and other consumer finance companies.

Do you think that's sort of where the thought process this is going to migrate for the entire industry? Or is this -- have you had dialogue around this with other companies? Or is this your -- just your sort of best thoughts about how to manage around CECL?

Ray Quinlan -- Chief Executive Officer

In this case, and of course all the CECL pieces are projections for everybody. And as you know from listening to other earnings calls, we are I think more engaged in the details associated with CECL and the impact on our particular franchise than most of our companies that are similar to ourselves. We have talked to people informally, and we have asked several advisors for their comments on both our disclosures on CECL, the impact associated with it going forward on capital, as well as the possibility of looking at it a particular way or a different way in keeping with the conclusion that we reached on the adjusted core earnings. And so I think a whole bunch of people have thought about CECL.

We think we're relatively close to the front end of that parade than the back of that parade. And so it's very hard for us to say what the industry standard will be going forward. But a key piece of things will be one, everybody getting over the CECL hump in the first quarter and then the regulators recognizing that they were successful in getting more equity into the regulatory regimen. Then the question will be how much is too much and what do we count? But I would say that those conversations are barely incipient at this point, and so I think it would be foolish of me to say what either the industry thinks or any other competitor.

Henry Coffey -- Wedbush Securities -- Analyst

And then if I've got this right, and I hope I have, it looks like your tax rate went down a lot in the quarter. Was that some specific item? And I got on the call late, so if you've addressed this already, I apologize.

Ray Quinlan -- Chief Executive Officer

OK. Yes, we've noted your tardiness. And so it is case that the tax changes in the quarter were idiosyncratic and catch-ups on a couple of tax situations that we had. And we'll return to our normal rate going forward unless something else were to come on the scene of which we are unaware.

Henry Coffey -- Wedbush Securities -- Analyst

And then the TDR, was that reserve adjustment in the June quarter? Or does that go on during the course of the year or how does it sort of flow into the numbers?

Ray Quinlan -- Chief Executive Officer

Yes, a couple things. One is since interest rates have started to decline, that phenomenon has been with us. And so there was $5 million, $10 million in the first quarter, but it was just covered by us. What we -- the drop that we had in the second quarter of the relevant interest rates was more than we had forecast at the end of first quarter.

And so one is, this has been around. Two is unless there's a change in the -- in our expectations, and we tend to use the forwards so that we don't have any crystal ball forecasting on this, so we want to be right down the middle of the lane. So unless there's a change in the forwards in a decline sense, we won't have any further adjustment for this type of thing.

Henry Coffey -- Wedbush Securities -- Analyst

Great. Thank you very much.

Operator

Your final question comes from the line of Ann Maysek with Rose Grove Capital.

Unknown speaker

Hi, good morning. Thanks for taking the question. Given the Fed's really public stance on encouraging the market transition from LIBOR, a lot of U.S. banks have recently redeemed LIBOR-based data preferred and then reissued them in either SOFR CMT form and using ARRC language when appropriate. Is this something Sallie Mae has considered? And then relatedly, how are you addressing the LIBOR issue on the asset side of the balance sheet?

Steven McGarry -- Chief Financial Officer

So thanks for the question, Ann. Regarding the transition to LIBOR, we continue to follow the progress that is or is not being made by, I guess they call it the ARRC group. We do have significant exposure to LIBOR on our balance sheet, on our variable rate loans, on our asset backed bonds, on our derivatives and of course, on the preferred that you are asking about. We will -- we've continued to change the language and the various securities that we issue as we issue them, and we are actually waiting to see what the group comes up with on the transition to before we take additional concrete actions.

Regarding that security, we don't intend to redeem it at this point in time.

Unknown speaker

OK, great. Thanks. Thanks for your help.

Operator

I will now turn the call back over for closing remarks.

Ray Quinlan -- Chief Executive Officer

OK. Well, thank you. And thank you for all for your questions and interest in our franchise. And in closing, I would just say that it's a pleasure to be able to talk to you all about this company, which has had terrific -- has had a great track record providing excellent returns on equity.

We do have controlled expenses, as you've seen with the efficiency ratio. Rational capital allocation going forward will be consistent with how we've conducted ourselves over the last year. Return of capital, there's been a bunch of questions on. Our motivations will be to get that on a transparent and crisp playing field and be able to forecast that going forward once the CECL piece is slightly behind us.

There is leverage in the franchise, as you can see with, costs growing less than one-third or growing about one-third of revenue. And we're fortunate to be able to be providing a service that is highly valued both by our student customers, as well as their families. And so as we continue forward in the tail end here of 2019, want to thank you for your interest and we'll talk to you next quarter. Take care.

Brian Cronin -- Vice President of Investor Relations

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

Thank you.

Duration: 60 minutes

Call participants:

Brian Cronin -- Vice President of Investor Relations

Ray Quinlan -- Chief Executive Officer

Steven McGarry -- Chief Financial Officer

Sanjay Sakhrani -- KBW -- Analyst

Michael Kaye -- Wells Fargo Securities -- Analyst

Moshe Orenbuch -- Credit Suisse -- Analyst

Arren Cyganovich -- Citi -- Analyst

Rick Shane -- J.P. Morgan -- Analyst

Vincent Caintic -- Stephens Inc. -- Analyst

Mark DeVries -- Barclays -- Analyst

Henry Coffey -- Wedbush Securities -- Analyst

Unknown speaker

More SLM analysis

All earnings call transcripts