Citizens Financial Group Inc

NYSE:CFG

Earnings Call Transcript 2020-Q2

[Operator]

Good morning, everyone, and welcome to the Citizens Financial Group's Second Quarter 2020 Earnings Conference Call. My name is Perkie, and I'll be your operator today. [Operator Instructions].

As a reminder, this event is being recorded. Now I will turn the call over to Doug Levy, Senior Vice President, Investor Relations. Doug, you may begin.

[Douglas Levy, SVP, IR]

Thank you, Perkie. Good morning. We're really pleased to have you join us. First, this morning, our Chairman and CEO, Bruce Van Saun; and CFO, John Woods, will provide an overview of our results, and we'll reference our presentation, which you can find at investor.citizensbank.com. Then we'll be happy to take questions.

Brendan Coughlin, Head of Consumer Banking; and Don McCree, Head of Commercial Banking, are also here to provide color. Our comments today will include forward-looking statements, which are subject to risks and uncertainties, and you should review the factors on Page 2 of the presentation that may cause our results to differ materially from expectations. We also use non-GAAP financial measures, so it's important to review our GAAP results on Page 3, and use the information about these measures and their reconciliation to GAAP in the appendix. And with that, I'll hand it over to Bruce.

[Bruce Van Saun, CEO & Chairman]

Thank you. Good morning, everyone, and thanks for joining our call. The second quarter posed unprecedented challenges given the impacts from the coronavirus and widespread disruption to people's lives and the economy. Once again, I am pleased that Citizens is rising to the occasion and delivering wealth for all stakeholders. We are taking great care of customers and colleagues while posting strong results that demonstrate the diversification and resilience of our business model. We also announced further commitments to diversity and inclusion, along with initiatives to promote racial equity and social justice.

Our financial performance in the second quarter featured tremendous revenue generation and strong profitability in our mortgage business. We made an important investment in acquiring Franklin American Mortgage company in May 2018 in order to gain scale and diversify origination channels in the business. In addition, we've made investments in talent, customer experience, and in digitizing and streamlining the business, which has positioned us well to capture the market opportunities we've seen since the middle of last year.

These strong results have been a balance to windward during the low rate environment and disruptions arising from the pandemic. Overall, our fees were up 28% year-on-year and 19% sequential quarter. With stable net interest income, total revenue was up 7% year-on-year and 6% sequential quarter. We did a good job on expenses, which resulted in 5.9% positive operating leverage year-on-year, a 54.9% underlying efficiency ratio and PPNR growth of 15% year-on-year.

Net deposit charge-offs as our credit costs in Q2, we had a record quarterly earnings of $1.14. As expected, however, we again built our credit reserves under CECL given the deterioration in the macroeconomic conditions since the close of the first quarter. Our ACL loans ratio is now 2.01%, and that's 2.09%, excluding PPP loans. In addition, we are selling a long-duration student loan portfolio, which free up additional reserves for reallocation. We feel we have good coverage now of the credit risk in both the consumer and commercial portfolios, though uncertainty on the path of economic recovery remains.

We have updated the granular information on credit portfolios, including some additional metrics in the appendix to our earnings presentation. We'll take a look. The strong PPNR generation and reduction in commercial line draws during the quarter helped improve our CET1 ratio to 9.6%, which is up from 9.4% in the first quarter. We had a very liquid balance sheet during the quarter, with average deposit growth of 12% sequential quarter, 8% spot. Our spot LDR at quarter end was 87.5% or 84.5%, excluding PPP loans. So overall, we have a very strong capital liquidity and funding position that allows us to use our balance sheet in support of our customers.

We continue to track well on all of our key strategic initiatives for 2020, and we've been working on refreshing our strategy to incorporate key trends and learnings from the crisis. We aim to take advantage of some of the opportunities we see to come out of the crisis well positioned for future growth.

And I hope you and your families are coping with the current challenges and remain healthy and safe. With that, let me turn it over to John for a thorough review of our financials. John?

[John Woods, Vice Chairman & CFO]

Thanks, Bruce, and good morning, everyone. Let's start with a brief overview of our headlines for the quarter. As Bruce said, this was an outstanding quarter for Citizens against a difficult operating backdrop. This allows us to head into the second half of the year with good momentum and excellent balance sheet strength. The resilience of the franchise is on display, as we generated $0.55 of EPS on an underlying basis. This was driven by record revenues and fee income given record mortgage fees, which offset headwinds in several other fee categories.

Net interest income was stable in the quarter, given strong loan growth, which offset a 22 basis point decline in margin. This was driven by lower rates and higher cash balances, but we did well in cutting deposit costs in half. We increased our allowance for credit losses to $2.5 billion. This translates to an ACL coverage ratio of 2.09%, ex PPP, up from 1.73% last quarter. We showed excellent balance sheet strength, ending the quarter with a stronger CET1 ratio of 9.6%, up 20 basis points from the quarter. Our liquidity ratio has also improved as we ended the quarter with an LDR of 84%, excluding PPP loans, and we remain in compliance with the LCR. Also, our tangible book value per share is over $32 at quarter end, up 4% compared with a year ago.

Now let me move to the highlights of our underlying results covered on Pages 4 and 5. Even in the midst of the COVID-19 pandemic and another strong reserve build, our results highlight the resilience of our diversified business model. Our EPS of $0.55 was down $0.41 year-over-year, but up $0.46 linked quarter. PPNR of $790 million was a record, up 15% year-over-year and 17% linked quarter. And in addition to another exceptional performance in mortgage banking, we also saw strong underlying performance in IRP and improvement in capital markets results.

Average loan growth was 6% in the quarter, reflecting PPP lending and the impact of the commercial line draws we saw last quarter, which benefited NII and helped to offset the impact of the more challenging rate environment. If we adjust for the sales, PPP and line draws, average loans were up 1% linked quarter.

Moving to Page 6. I'll cover net interest income, which held up quite well despite a lower margin. Net interest income was stable linked quarter as the benefit of 8% interest-earning asset growth and accrued funding costs was offset by the impact of lower rates. Net interest margin decreased 22 basis points linked quarter as the impact of lower rates and higher cash balances was partially offset by lower deposit costs and outsized growth in DDA and other lower-cost deposits. About 6 basis points of the margin decline related to higher cash balances in the quarter, given strong deposit growth, as consumers and small businesses benefited from government stimulus and corporate clients-built liquidity.

We were especially pleased with our progress on deposit costs, which we drove down 37 basis points during the quarter, a more than 50% decline. Our total interest-bearing deposit costs were 48 basis points at the end of the quarter. That compares to 34 basis points back in 3Q '15 at the end of the last ZIRP period. So clearly, we have a near-term opportunity to continue to reduce these costs.

Moving to Page 7. I'll discuss fees, which really shows the benefit from the work we've done to build capabilities and diversify our business. Noninterest income was a record, up 19% on a linked-quarter basis and 28% year-over-year. Record results in mortgage banking were partially offset by continued headwinds related to COVID-19 and other fee categories. On a sequential basis, mortgage banking fees increased by 74% to $276 million, reflecting continued strong refi and HELOC volumes, and record high gain-on-sale margins in particular. Capital Market fees of $61 million increased $18 million from first quarter, reflecting strong DCM activity and a $13 million mark-to-market recovery on loan trading assets.

Foreign exchange and interest rate product revenues increased 5% linked quarter before the impact of CVA. Interest rate product sales led the way as clients repositioned for a lower rate environment. CVA improvement was $8 million in the quarter. Trust and investment services fees were lower by $8 million linked quarter given the rate environment and the effect of the equity market decline on managed money revenue. The service charges in card fee categories were down significantly compared to first quarter, reflecting the full quarter impact of the shutdown and impacts from stimulus money to customers.

On a positive note, we see debit card activity roughly back to pre-pandemic level. And credit card activity in June, only down about 10% compared to last year, a significant improvement from the over 30% declines we saw in early April.

Turning to Page 8. Underlying noninterest expense declined 2% linked quarter, largely driven by CECL impacts in Q1 on salaries and employee benefits. Salaries and employee benefits declined $30 million or 6% linked quarter, largely reflecting seasonally lower payroll taxes. Equipment and software expense and outside services were higher linked quarter and reflects increased technology spend and investments in growth initiatives.

Next, let's discuss loan trends on Page 9. Average core loans were up 7% linked quarter, primarily driven by the full quarter impact of the commercial line draws at the end of the first quarter and the $4.7 billion of PPP lending to our small business customers. Before the impact of loan sales, line draws and PPP loans, core commercial loan growth was up approximately 1% linked quarter. The $7.2 billion of post-COVID commercial line draws in March have been substantially repaid and were down to $1.8 billion by the end of the second quarter. Overall utilization is down to approximately 40% from 50% at the end of the first quarter.

Core retail loans on a linked-quarter basis were stable with growth in education and other retail, offset by lower home equity balances and the transfer of approximately $900 million of education loans held for sale. We are building an originate-to-distribute model for our consumer assets, which will generate fee revenue and increase our balance sheet flexibility over time.

Moving to Page 10. Deposit growth was exceptionally strong in the quarter. We saw a robust average deposit growth of 12% linked quarter and 15% year-over-year, outpacing loan growth and driving our average LDR down to 89%, excluding PPP, as consumers and small businesses benefited from government stimulus and clients-built liquidity. These strong deposit growth came in lower cost categories, with average DDA growth of 25% on a linked-quarter basis and 33% year-over-year.

We continue to aggressively execute our deposit playbook to manage down our deposit costs across all channels. We were able to cut our interest-bearing deposit costs by roughly half this quarter, down 46 basis points to 48 basis points, and down 82 basis points year-over-year.

Let's move to Page 11 and cover credit. We continue to assess the impact of the COVID-19 pandemic and are closely monitoring the portfolio for areas of potential risk. That said, portfolio performance is progressing largely in line with our expectations, but with a somewhat more adverse macro backdrop than we saw at the end of the first quarter.

Net charge-offs were stable at 46 basis points linked quarter as increases in commercial were partially offset by improvement in retail, reflecting the impact of forbearance. Nonperforming loans increased 27% linked quarter, driven by a $192 million increase in commercial, reflecting COVID lockdown impacts, and an $18 million increase in retail. The nonperforming loan ratio of 79 basis points increased 18 basis points linked quarter and 17 basis points year-over-year. However, in spite of this increase, the nonaccrual coverage ratio remained strong at 255% at June 30.

We increased our CECL credit reserve coverage ratio from 1.73% in 1Q to 2.09% in 2Q, excluding PPP loans. This 36 basis point increase was primarily driven by a net reserve build of $317 million. In addition, approximately $100 million of reserves associated with a planned sale of student loans were reallocated to the remaining loan portfolio. In effect, the reserve build was $417 million or 90% of the Q1 goal.

On Page 12, we provide detail on customer forbearance and the PPP lending program. We continue to work directly with our customers to assist them through these challenging times and have seen encouraging trends. The average FICO score of our retail forbearance customers remains high at 725, and approximately 93% of these loans were current when they entered forbearance. We also continue to work proactively with our commercial clients. Grantee release were needed in the form of covenant modifications will allow for PPP applications as well as granting selective temporary relief on principal and interest payment.

I'm also pleased to say that through June 30, we helped our customers receive $4.7 billion in PPP loans, which has allowed us to help support over 540,000 jobs. 84% of loans made were below $100,000.

Moving to Page 13 to discuss our CECL methodology and reserves. We have summarized the key aspects of our macroeconomic scenario, which is a foundational element of the CECL reserve estimate. At quarter end, we elected to use the May 13 Moody's baseline as our base scenario. Similar to last quarter, given the uncertainty of the continued economic outlook, we also considered other Moody's and internal scenarios. In general, our aggregate economic scenario is more severe than that used in the 1Q. It assumes the steep drop in GDP in 2Q, which is followed by a gradual recovery in the second half of the year and into 2021. If this scenario plays out, provision requirements over the second half of 2020 should be more reflective of net loan growth and incorporate a smaller build. However, if the pandemic impacts are deeper or if it takes longer for the economy to recover, or government programs are less effective than we expect, then we could require further additions to reserve levels.

On Page 14, as I mentioned earlier, we feel well positioned to manage through the current environment with strong capital and liquidity positions. Our CET1 ratio improved to 9.6%, up 20 basis points linked quarter given our strong results and a reduction in risk-weighted assets. Additionally, during the quarter, we issued $400 million of Tier 1 qualifying preferred stock, which, in combination with the increase in CET1, drove a 40 basis point increase in Tier 1 capital. Strong deposit growth outpaced loan growth, which improved our liquidity metrics and drove the spot LDR, excluding PPP loans, down to 84%.

Turning to Page 15. Let's look at reserves and capital versus stress losses. Our ACL of $2.5 billion represents a very strong 52% of our modeled losses using the Fed scenario, and is now 38% of the stress losses in the Fed's 2020 DFAST. In addition, when adding excess capital above our preliminary SCB of $3.4 billion to our ACL, the resulting $5.9 billion is 120% of our estimates and 88% of the Fed loss estimates. These levels are further fortified by the additional coverage from the PPNR we generate. On average, we've generated approximately 35 basis points of CET1 capacity per quarter over the last 6 quarters.

On Page 16, we show a summary of the Fed's stress test results. The Fed estimated our PPNR at 2.3% of average assets, which is well below the peer median of 3.3%. We believe this ignores the steady and significant progress we have made to improve our PPNR since the IPO. For example, our PPNR to assets for 2019 has improved by approximately 37% since the IPO to 3.7%. Importantly, this compares to a stable 3.7% in actual PPNR to assets during the first 6 months of real-life stress in 2020. The Fed's estimate of our credit losses at 5.6% was right on top of the peer median and down from 6.1% in 2018. However, our estimated company-run severely adverse credit loss rate of 4.2% is significantly lower. We believe that the Fed's modeled results and the 3.4% preliminary SCB is elevated above what our business model would imply. As such, and as we indicated in our CCAR release in June, we have submitted a request to the Fed to reconsider our preliminary SCB.

On Page 17, I want to highlight some exciting things that are happening across the company. While we are first and foremost focused on helping our clients, we are looking forward and continue to work towards building a better company. We continue to execute on the transformational TOP program and are making steady progress towards our targets. Planning is underway to add significant new transformation initiatives, including the end-to-end digitization of customer interactions and operations as well as other initiatives to adapt to the post-COVID-19 environment. We are also moving forward with our major strategic revenue initiatives while considering new opportunities arising from the current environment in an effort to drive higher revenue growth coming out of the crisis.

Moving to Page 18. We provide some commentary on how key categories are shaping up for full year 2020 compared to the prior year. We expect net interest income to be broadly stable as loan growth is offset by a meaningful decrease in NIM due to lower rates. Noninterest income is expected to be meaningfully driven by the exceptionally strong results in mortgage, which more than offset the weakness in other fee categories related to COVID-19. We expect several key fee categories to benefit from a return to more normal activity levels in the second half, which will help cushion a moderation in mortgage revenue.

Noninterest expense is expected to be up modestly, particularly given higher compensation tied to stronger mortgage production and impacts from COVID-19, which includes government lending programs and customer relief efforts. Provision expense is the greatest potential for variability and remains dependent on the path of the recovery. We expect solid loan growth, driven by the impact of higher line draws in commercial during the first half and government programs like PPP as well as increased demand in education and merchant financing.

Our capital position remains robust, with our regulatory capital ratios expected to improve further over the remainder of the year, driven by net income growth, a moderation in RWA growth in the second half and the suspension of our buybacks through year-end. Looking forward, we expect to remain well capitalized and feel confident we can continue to maintain the dividend at current level.

Now let's move to Page 19 for some high-level commentary on the third quarter. We expect NII to be up modestly, reflecting PPP benefit on NIM. Excluding PPP loans, loan growth is projected to be down modestly due to the full quarter impact of a decline in commercial loan in line with utilization in the second quarter. Ex PPP, the NIM is expected to be broadly stable with the benefit of lower deposit costs being offset by ongoing rate headwinds.

Fee income is expected to be down in the mid- to high single-digit range, reflecting lower mortgage banking fees from record levels, partially offset by recovery in other fee categories. Noninterest expense is expected to be up in the low single-digit range, reflecting higher origination-related costs levels in the mortgage business. We currently expect a smaller reserve build, though provision expense will be highly dependent on an updated view of the economic recovery and portfolio performance. Finally, we expect average loans to be down in the low single digits given the pay down in commercial line draws during the second quarter. Excluding the impact of line draws, PPP and loan sales, we expect loan growth to be broadly stable.

To sum up, our profitability, capital and liquidity position remains strong, and we are delivering well on our key initiatives for stakeholders. Now let me turn it back to Bruce.

[Bruce Van Saun, CEO & Chairman]

Thank you, John. Operator, let's open it up for Q&A.

[Operator]

[Operator Instructions]. And our first question comes from the line of Scott Siefers with Piper Sandler.

[Robert Siefers, Piper Sandler & Co.]

I guess just sort of a top level question on reserving and adequacy and methodology. There's, of course, so many moving parts now with the vast uncertainty and then the Fed sort of through a -- rends into things with the -- their sort of earnings sufficiency test. But just as you look at the number, 2% is certainly very, very strong and up big from the first quarter. But as you think about things, what makes 2% the right number versus, say, some who have upped the ante to like 2.5%, given your own loan mix outlook, et cetera. How do you arrive at that conclusion, given all the moving parts?

[Bruce Van Saun, CEO & Chairman]

Sure. Let me start, and then I'll flip it to John. But actually, we look at it, Scott, first off, ex the PPP loans, so it's really 2.09%, and the folks you're referencing that are in the mid-2s typically have very sizable card businesses, and we have a de minimis card business. And so, if you strip out the card from the other banks, the reserves on their card book, they're generally around 2%. So, we feel that we're right in the pack with everybody else on that basis.

So you can't just look at a direct kind of high-level number, you have to actually peel back and look at the individual portfolio and see what the coverage is. And if you look at coverage by each of our consumer portfolios and then by our CRE and our C&I, I think we feel quite confident that we have adequate reserves for the scenario that we've laid out. And that kind of is the key. You have to go through this process of choosing a base scenario, contrasting with some other scenarios. Thinking about the impact of the government stimulus and how long it lasts and what the benefits are, think about forbearance and what benefits that's providing. So, there's a lot of assumptions that go into arriving at a new scenario, and we went through that whole process, and this is what we come up with.

If it turns out down the road that the reopening is slow and that the scenario extends the recovery from what we assumed this time, there's always the chance that we have to take a further look at reserve build. We did include in the slide deck, on Slide 13, just a little table that says if we had another -- if we wanted to use up 10 basis points of CET1, we gained another 24 basis points in ACL coverage. So, the 2.09%, we go to 2.33%. So, I think the broad point there is that we feel we have very strong levels of T1. We feel we have a very robust ACL ratio. We're also pleased that our PPNR is staying strong and resilient through this stress period. And so, it's all those elements that give you confidence that you have ample capital to absorb any credit losses. With that, I'll flip it to John.

[John Woods, Vice Chairman & CFO]

Yes. I think you captured it, Bruce. I would just maybe add a couple of points. We get worse in our scenario, and I think that was prudent quarter-over-quarter. But we have -- we did see that the consumer ends up being extremely resilient, and so just given the impact of all the stimulus and forbearance, so consumer has really been performing well. So that's been factored into our numbers.

From a commercial standpoint, you'll notice in our materials that we increased that amount of reserves significantly. We took that up from 1.2%, up to 2.16%. So, we think that, that was prudent, and we think that's an appropriate way to express where we see things at this stage, and then as Bruce mentioned. And you sort of have to look at reserves in concert with capital. Just in the CECL world, our number is 2.09%. When you compare that to others at ex card and then you layer in our capital at 9.6%, when we look at that in the totality, we feel this is an extremely strong approach to reserving and managing our capital.

[Robert Siefers, Piper Sandler & Co.]

All right. That's perfect. And then if I can slip one more in. You guys have always done a very nice job of sort of getting ahead of things with your TOP programs, always having something in the tank to support your PPNR. And of course, the current TOP program is much bigger than past ones. But just given the change from what we would have contemplated, say a year ago when this was first put out there, is there any chance to sort of revisit things, maybe get even a bit more aggressive on cost savings or things like that?

[Bruce Van Saun, CEO & Chairman]

Yes. So we did indicate in the materials that we are focused on some new ideas based on things we're seeing through the whole pandemic period, particularly the increased use of digital channels and what that portends in terms of accelerating the move to digital across our businesses. And we think that offers some meaningful upside in terms of efficiencies across really distribution, operations, and technology. So, we're in the process of scoping that out.

And there's other aspects, use of virtual advisory and other things, and how we work in terms of how much remote and how much office space we need. There's a number of things that we're looking at that I think offer some big potential for savings down the road. I did mention last quarter, I think, that we have this effort called -- so-called war on paper that Brendan here is championing. There's a lot of paper in the consumer bank. And so numerous opportunities, I think, to increase the scope of what we're doing.

[John Woods, Vice Chairman & CFO]

Yes. I might add that we've kind of put all of our TOP 6 initiatives through the lens of a post-COVID world and where we're headed. And we validated that not only we were on the right path, but we may even be accelerating on all of the things that we're doing with next-gen tech, and converting to an agile workplace and ways of working and then broadening it out. As Bruce mentioned, we see new opportunities in the digitization phase.

[Operator]

And our next question comes from the line of Ken Usdin with Jefferies.

[Kenneth Usdin, Jefferies]

I was wondering if you could just give us a little bit more detail on some of those fee drivers on the mortgage business, outstanding again. And you talked about expectations for it to come down a little bit, but just what you're looking out for in terms of those balance between gain on sale and origination? And then also, you mentioned the expected improvement in some of the other line items, but if you could really just walk us through service charges and what you're seeing as far as inactivity as that -- how that came through the quarter and how you're expecting it to react?

[Bruce Van Saun, CEO & Chairman]

Yes. I'll go ahead and start, and maybe Brendan may want to chime in here. But just from a mortgage banking standpoint, a phenomenal, phenomenal quarter. And really the second quarter, the story was about margins. I mean when you look at how we recognized revenue, it's basically on hole through adjusted locks. And those were strong, I mean, at elevated levels, but really not too different than first quarter. Really, it was about margins that were basically doubled from 1Q, and that's as high as it's ever been, certainly record for us and possibly into the industry. So that's what was happening in the second quarter. When you look out into the third quarter, we still see strong locks, possibly even higher, who knows, we'll see. And -- but margins coming down off of record levels. I mean we suspect that margins will hang in, possibly above 1Q levels, but it would be hard to predict that we would be able to sustain the margins that we saw in 2Q. So that's why we expect to have another strong quarter for mortgage, but maybe tempered and moderated from where we were in 2Q from a mortgage standpoint.

And then in the service fees and card area, we're seeing at the end of the quarter some better uptake in terms of activity. And maybe I'll just turn that over to Brendan to add some color there.

[Brendan Coughlin, Head, Consumer Banking]

Yes. I think you summarized mortgage well, John, so I won't add much there. On service charges, the headwind, what we have is around NSFR, and you can see in our results that enormous amount of growth in DDA from the stimulus and a lot of that money is still sticking around, which has a direct and adverse impact with NFS. So we believe that would be largely temporary, but we'll have some persistent pain to the rest of the year as that balance parking sticks around and ends up burning off through the year and into next year.

On card fees, debit card, as John pointed out in his prepared remarks, is basically back to almost par from year-over-year, and credit card is getting there. We've got a little bit of pain in credit card with some slowdown, intentionally, on balance transfer initiatives just given the state of credit, while we make sure we've got our arms around it before we turn back on any growth engine in the card books. That's a lever we have once the economy stabilizes, and we feel confident enough to turn it back on in the second half of the year.

And then I would say, just lastly, on wealth fees, we had a bit of a reduction in wealth, just given the market pullback with our managed money book, which has started to recover, as you know. So we see some tailwinds in the second half of the year with that. And sales have all but dried up in Q2 with our branches in appointment-only mode, and we've seen a real material pickup in sales, both managed money and transactional sales. So we do expect a bit of a rebound in wealth fees in 3Q and the second half of the year.

[Kenneth Usdin, Jefferies]

Got it. And just one question on deposits. The growth at period end was almost as good as it was on average. How do you assess the stickiness of that? And what's coming in existing customers versus new customers?

[John Woods, Vice Chairman & CFO]

Yes. I mean yes, the year -- exactly, spot growth in the second quarter was quite strong at 9% and year-over-year, 10%. So you're right. It's like it's been hanging in extremely well. They're down a bit from the average balance growth of 12%, but hanging in quite nicely.

I think what we're trying to do is to monitor the behavior of consumer bank and our commercial customers. So on the consumer side, we had stimulus, PPP, balance parking, tax deferral, unemployment, I mean all of those forces. So basically, we expect that to moderate over the remaining part of the year, but it really depends upon the level of stimulus and what's happening with spending the PPP dollars. And then on the commercial side, we have also PPP and line draws, right? And so those line draws have come down by a lot. Well, they continue to run off. So I suspect that you will see that moderate through the rest of the year, highly dependent though on further fiscal stimulus and customer behavior.

[Operator]

And our next question comes from the line of Ken Zerbe with Morgan Stanley.

[Kenneth Zerbe, Morgan Stanley]

I guess, why don't we start off in terms of the student loan portfolio, the $900 million that you sold. Can you just talk about the rationale for doing that? And also, does this represent any kind of change in terms of how you're approaching the business?

[John Woods, Vice Chairman & CFO]

So I want to go ahead and start off here. I think that really the broader context is really balance sheet efficiency and balance sheet optimization. So what we've been endeavoring to do -- certainly in the mortgage business, we have a pretty highly tuned originate-to-distribute model, and we also have loans and portfolio of mortgages on balance sheet. And that is the -- we believe that's the appropriate balance, is to have some balance sheet availability for consumer assets and some ability to distribute that paper.

So we're doing that in the mortgage space, of course, and many companies are as well. But we're looking to be able to migrate to that kind of balance over time in the student book, in the auto book, et cetera. And so student is an area where we have a pretty solid origination capacity and capability. And we have identified an opportunity to get to begin to migrate into the originate-to-distribute arena, if you will, with student. And therefore, we -- with that as a good pilot transaction, we'll be better equipped to balance portfolio lending and sales going forward.

And importantly, we're keeping the customer relationship in this sale. So it's truly just balance sheet optimization while maintaining the strategic customer relationship. So we think it's a good approach. So with that, I'll maybe see if Brendan wants to add.

[Brendan Coughlin, Head, Consumer Banking]

You covered it well, as usual. The only thing I would add is in the student portfolio, there's multiple different products under there. One is our student loan refinance product. And then the in-school portfolio as the balance as we move to held for sale, which is significantly more CECL-intensive than the student loan refi portfolio. So -- and both of which we think are still very distinctive for us in customer acquisition, even you start to see some market pull back with some of the bigger lenders curtailing their originations over the last quarter. So we still do believe we're very bullish on it. We think we've got some white space here to acquire customers and optimize the use of the balance sheet as we can and hopefully gain on sale over time with our position.

[Bruce Van Saun, CEO & Chairman]

Yes. I would just add, too, that this portfolio that we've moved to held for sale is very attractive from a yield standpoint. So we think we can get a good price, ultimately, an execution on this, and at the same time, free up the reserves that we were holding against it. So in effect, we've taken about $100 million, and we're reallocating that to other portfolios, which augments the actual reserve build that we took. So from a timely standpoint, all the things John said about strategically just makes sense, but it helps us boost our CET1 ratio and it helps us move those reserves elsewhere to other portfolios and continue to bolster our reserve coverage ratios.

[Kenneth Zerbe, Morgan Stanley]

All right. Perfect. And then just maybe a second question for you. Just to put you on the spot a little bit about provision expense, and I will say that it's really encouraging to see provision expense might be very much under control this quarter. But when we think about provision expense going forward, given your guidance, I know it's really hard to pin down numbers, and I'm not trying to necessarily, but when you talk about less reserve build and higher NCOs, I think there's a scenario where your provision expense could be higher from second quarter, it could be lower than second quarter, it could be meaningfully lower than the second quarter and still fall within your guidance. Given what you know today, is there -- I mean, would you expect that provision expense to be meaningfully lower or some kind of in the middle?

[John Woods, Vice Chairman & CFO]

Yes. I'll go ahead and start off there. I mean as we mentioned in our outlook, this is just highly dependent upon a variety of factors, right? I mean we're -- in kind of May, we fell a certain way. In June, we fell a little differently. The things -- April, we felt really a lot worse. So things are going to really play out over the next 60 to 90 days in terms of what's going on with the rate of infection around the United States, the rate of closures. And I think you've got to look at that. You've got to look at how our portfolio performs. Will the consumer remain resilient? Will our commercial outlook remain as expected? I mean there's just so many factors. I just think that there's a lot more that we don't know than what we do know about what our provision could actually be when we were sitting here in September. We just feel like, more broadly, if things play out the way we expect, that provisions going forward would naturally, given the way CECL works, be more closely tied to loan growth than large builds going forward. So I mean, I just -- you kind of throw all that together, and it's just very difficult to indicate what that expense will be.

[Bruce Van Saun, CEO & Chairman]

I would just also add, it's kind of a -- look, the economy is recovering, reopenings are occurring. The trend line is up, but it's really in a sawtooth pattern. So you kind of go from month-to-month, and you we feel -- starting to feel pretty good, and then whoops, you're not feeling quite as good.

But we put away a lot in the first half of the year and I think we're cautiously optimistic that the reopening process, even though we're seeing kind of the sawtooth at the moment, is still moving -- progressing in a positive direction in the second half of the year. So I think that would argue, unless something else happens, that kind of the trend line of the provision was biggest in the first quarter. It was smaller in the second quarter. You could see a scenario where that continues on into Q3 and then into Q4 at this point.

[Operator]

And our next question comes from the line of Erika Najarian with Bank of America Merrill Lynch.

[Erika Najarian, Bank of America Merrill Lynch]

My first question is for you, Bruce. Clearly, all the work that you did to diversify your revenue stream is paying off in spades, GAAP EPS going from $0.03 last quarter to $0.53 this quarter. And I'm setting up the question because clearly what surprised folks on the DFAST results for the industry is the introduction of an income test. And so as you think about the earnings power going forward and your comments on future provisions, it would be great to hear from you in terms of what you believe the dividend sustainability is if the Fed continues to extend this income test beyond the third quarter.

[Bruce Van Saun, CEO & Chairman]

Sure. And let me just first pick up on the efforts to diversify the business model, but we aim to be good stewards of capital. And I think all of those 5 kind of fee-based acquisitions that we've done, 3 in the M&A space, Franklin in the mortgage space and then Clarfeld in the wealth space, have been really terrific. It all met our expectations and are helping to bolster that fee income, which is really helpful in this type of environment. So we'll continue to look for those opportunities. We've also made a lot of organic investment in people and capabilities and technologies to bolster those areas as well.

When I look forward -- when we put out the commentary around the Fed's DFAST results, we said we're quite confident that we can continue to sustain this level of dividend through the year. And so if you look at the guidance that we're giving, the results we have today, certainly, the third quarter outlook and then the broader guidance we're giving for the full year, I think you can adjust your models and you'll see that, that's the case, that there's going to be ample headroom to be able to sustain the dividend where it is. We won't comment yet on 2021 because, clearly, we tend to recalibrate and offer guidance at the outset of the year on our January call. But again, I think the resilience that you're seeing, there's no reason that, that shouldn't also continue into next year.

[Erika Najarian, Bank of America Merrill Lynch]

Got it. And the second question is -- and when you scoured the Q before the call, I'm wondering if you could give us a little bit more detail on the loss-sharing agreement. Clearly, it's part of the disconnect in terms of co-run losses versus Fed-DFAST losses. Maybe it's also part of why you're getting the questions on whether or not a 2.09% allowance is adequate. So wondering if you could share with us in a little bit more detail how much of your consumer loans does have this -- a loss-share agreement, and when the charge-offs come, how does the loss-share agreement work and protect Citizens?

[John Woods, Vice Chairman & CFO]

I'm going to start off and maybe Brendan, what you're you going to add here, Erika. So just the point that we're making and that we've observed is that -- with the Fed is that in the card space, there are revenue and loss-sharing arrangements that are maybe not predominant, but certainly exist. And the Fed has, in the last round of DFAST, recognized that, that -- that they exist, and has begun to take specific account for that in their projections of losses in card.

And so generally, the way to think about it is that we have a very similar arrangement, a revenue loss-sharing arrangement, with respect to our merchant finance activities with varying terms, et cetera. And really our point is that those need to be recognized as well, even though these are not technically card assets. So that's the main message on that. And maybe I'll turn it over to Brendan to answer it for you, to react to the other points.

[Bruce Van Saun, CEO & Chairman]

Let me just chime in to that, John, though. But if you have the loss sharing, then obviously, the stressed losses of looking at individuals in a downturn flips to looking at your counterparty's ability to pay and cover the loss sharing. And the marquee partner that we have is Apple, which I think, last time I checked, has pretty damn good credit. So if you make that substitution, that would be very beneficial to the loss picture, and there's other arrangements like that. So that's just one element.

We actually have arguments that we put forth on PPNR. We have arguments even beyond that on the credit side. But anyway, we've commenced that process and we think that the Fed is -- we commend the Fed for actually opening this up and allowing a dialogue around whether these results are being modeled accurately because they really matter today. Before it was more of a pass/fail to how are you doing, but now in the SCB construct, it has more bite. And so even though we're fine, we'd like to kind of bring our numbers back to be more representative of a traditional regional bank, which is what we are. And so there's a number of factors that are causing us to be elevated. And the Fed has started to deliberate, and we've submitted our materials, and I think we'll get a fair hearing on these issues. Brendan, do you want to add anything?

[Brendan Coughlin, Head, Consumer Banking]

Yes. The only thing I would add is we don't disclose all the ins and outs of the arrangement partner by partner. But if you look at how it would stress, it's an unsecured asset, so if left without the loss sharing, you'd surmise that the merchant partnership should stress like a credit card. When you add them on, we project it to stress like a secured asset or even better, in some cases, significantly better. And then as I look at it independent from the loss-sharing arrangement, that portfolio in real terms not picked up by the Fed yet, early innings here on forbearance, but that portfolio and the merchant portfolio is actually in single digits basis points and forbearance, almost nil.

And the delinquency has been extraordinarily stable. And so it's performing as if it's a super prime asset without really a recession going on around it right now, which is incredibly a good early signal. We think a lot were driven by the innovation and the customer experience design and how the automatic payments are set up. So independent of the dialogue around the Fed, results of the underlying asset, we believe, is really well under control and performing quite well.

[Operator]

And our next question comes from the line of Gerard Cassidy with RBC.

[Gerard Cassidy, RBC Capital Markets]

John, I have a couple of questions around the forbearance numbers. And I apologize if you've already mentioned this. What percentage of the loans in forbearance are making payments? I know it's different by category, but I don't know if you've had any color there. And then second, do you have a sense yet from the Fed when their policies supporting forbearance may change, requiring banks like your own to have to reclassify the loans into different category, downgrade them and put more capital against that?

[John Woods, Vice Chairman & CFO]

Yes. I'll cover those points, Gerard. So I mean, the payment data is coming in, and we haven't really talked about it overall. But just category by category, I mean, big picture, we're seeing north of 50% in a couple of categories in terms of payments. When you think about a payment made within the last 60 days, you could talk about a resi mortgage portfolio being north of 50%, same with our card book. You can see large percentages in some of our other categories like auto and home equity making just short of 50% as well in terms of having made a payment, even while in forbearance in the last 60 days. So that's actually an extremely good story, and it could be just kind of a convenience for certain customers that -- just as a safety net, and they, in certain respects, continue to make payments. So that's that.

As it relates to -- really, there's an intersection of the regulators, the CARES Act and FASB as it relates to how we account for forbearance. But, by and large, there's an expectation that there's a full 180 days that would be permissible before we then bring it back to starting the clock on nonaccrual, and nonaccrual basically runs approximately 120 days thereafter. So that's one of the reasons why this is getting pushed out a little bit on the consumer side. Lots of stimulus. We're keeping an eye on round 2 of extension requests, and we're -- and a lot of the -- rubber will hit the road once they become off forbearance, which customers are starting to come off forbearance now. And then the clock starts ticking as it relates to delinquency statistics and nonaccrual, which are all pushed out into, frankly, early 2021 as we see the calendar and how it all shakes out, given that this thing started in late March, early April.

[Gerard Cassidy, RBC Capital Markets]

John, just a technical question on the forbearance. Are all the loans accruing interest, even the ones that are not making payments?

[John Woods, Vice Chairman & CFO]

We are accruing, but we are also putting up reserves against those accruals separately to be prudent. As it relates to -- so we don't want to put up a bunch of accrued interest receivable that later on has to get written off. So we're one quarter into this, and we have a separate reserve outside of CECL to just kind of prudently ensure that, that accrues versus this receivable number doesn't get too large, and we'll continue to do that going forward.

[Gerard Cassidy, RBC Capital Markets]

Very good. And then just quickly, Bruce, what's the biggest, in your interpretation of the federal reserves, results for the CCAR and DFAST numbers versus the -- your internal models, where do you see the big glaring differences? And are you hopeful or confident that you could -- maybe could put there to think your guidance win?

[Bruce Van Saun, CEO & Chairman]

Well, I'd say the biggest one that we've referenced over the years has been how they model PPNR, which if you -- you have to view us in a historical context, where we were owned by RBS, which had its challenges and Citizens was forced to shrink its balance sheet, which, in a high fixed-cost base, causes profitability to fall dramatically. And so when they were picking up data, they picked up -- initially, they overweighted that period right after the Great Recession, which really harmed us. And the banks who've got TARP actually got a benefit that when you play it forward and look at -- when you get into a stress environment the next time, we wouldn't be forced to shrink our balance sheet. We've made a lot of improvements to our profitability, and the other banks wouldn't get that TARP benefit. So it was kind of a double whammy.

We had some banks with inflated estimates on PPNR and we had decided negative one. The Fed announced that they have retriggered their model and are now putting more weight on recent periods, but we think they're still picking up some of that past period. And so our principal argument would be, for us, given the journey that we've been on, given all the improvements, we've releveraged the balance sheet, we've invested in our fee-based businesses, we've done acquisitions like Franklin, that it's not providing an accurate picture of who we are. And that's worth a lot of basis points in terms of the kind of peak-to-trough drawdown and the impact it has on SCB. And then further, the things that John mentioned, there are some specific things on credit like the loss sharing on these merchant portfolios that aren't getting picked up as well. So I can't tell you -- I think they're listening, and I think we put good arguments forward. I think if truth and justice prevail, we should get something. But to try to say today whether I'm confident or not, I said I'd just say I don't know. It's too early to tell.

[Operator]

And our next question comes from the line of Matt O'Connor with Deutsche Bank.

[Matthew O'Connor, Deutsche Bank]

Bruce, in your prepared remarks, you talked about taking advantage of opportunities out there. And obviously, they do not mortgage now. And you mentioned some efforts to accelerate digital and look at the cost base. But is there anything else that you were kind of referring to when you were making that comment, whether it's organic acceleration of some growth efforts or looking from an inorganic perspective, just wanted to see if you could elaborate on that comment?

[Bruce Van Saun, CEO & Chairman]

Sure. I'll start. Maybe John can add to it. But one of the things, Matt, that we've been very focused on is our -- we call them our strategic initiatives. We did a very deep refresh of our strategy to try to look at where do we see trends and opportunities in the marketplace that maybe aren't fully serviced by other competitors. And where do we have capabilities and strengths where we would have a "right to win". And we came up with 3 significant initiatives. One was our national digital bank is to try to take Citizens Access to the next level. The second was to leverage what we've done in the merchant space to have new offerings for merchants, including also offerings that go direct to consumers to give them financing flexibility and choices. And the third was to build out a small business in lower middle market platform rich with information and services that we think those businesses would really benefit from. And coming through the pandemic, we've had to refresh those and say, are they all -- are they -- or are all 3 full systems go? Do they make sense? And are there things that we should do to even push harder on some of these things? And I'd say there, Matt, the good news is that every one of those things is hitting the target. We think there's still a very big opportunity there. I think we can probably get more out of the digital bank than initially anticipated, probably get more out of merchant finance than anticipated.

I think the one that's had to stay in shackles a little bit has been the business bank initiative just because of all the effort and focus around PPP. And that market is in a little bit disarray. So we've had to pull a lot of our people to just stay focused on the PPP program and on now the forgiveness aspect of that. So that will be a little bit of a delayed start, but we still like the potential there.

So I'd say those are the real unique opportunities that we see. We do think like virtual wealth advice is another one that we're looking hard at. We've been able to operate well from a remote standpoint through this period. We do have an integrated digital and banking platform that, quite frankly, hasn't had a lot of take-up. But if we kind of marry that and play around with the formats and potentially have the robo at kind of digital service but augmented with some folks with gray hair to provide advice at a different price point, that could be an interesting proposition as well. John, anything you want to add to that?

[John Woods, Vice Chairman & CFO]

Yes. I mean I'd say that those were well stated. And I just would reiterate our enterprise programs in TOP 6 and BSO, which are opportunities for us to continue to drive value. And then as Bruce mentioned earlier, we've done well and we've been disciplined with respect to our fee-based acquisitions. And we continue to have interest in filling in gaps and diversifying that fee revenue profile going forward. And so that's another thing that we remain interested in pursuing.

[Matthew O'Connor, Deutsche Bank]

And any updated thoughts on potential bank deals? I mean it might take a little while to play out, but obviously, you're sitting here with a lot of capital. You issued a little bit of preferred and generating good PPNR as you mentioned. So some banks out there are kind of getting ready for some deals that may emerge over the next several quarters, and wondering what your thoughts are on that.

[Bruce Van Saun, CEO & Chairman]

Yes. Look, I don't really see us making any initiatives in that direction for a while. So at this point, I think just following through on all the change programs we have and the investments that we're making is going to be our area of focus. And if we keep delivering here, if we have a good credit outcome, then we should benefit and be better positioned down the road potentially to do things like that, but certainly not in the near-term burner at this point.

[Operator]

And our next question comes from the line of Saul Martinez with UBS.

[Saul Martinez, UBS Investment Bank]

So, John, can you give us more color on what PPP assumptions are embedded into your NII guide for the third quarter and for full year? Because you did say a significant amount forgiven in the third and fourth quarter, which seems a little bit earlier than maybe some of your peers have indicated who were suggesting it's going to be a little bit more backloaded. But if you can help us understand maybe some of the assumptions for overall forgiveness, timing, and just help us frame this and size up the potential benefits. And how much of your NII is being buttressed by PPP and how much is sort of being driven by core NII growth?

[John Woods, Vice Chairman & CFO]

Yes, sure. I'll throw a few thoughts out, and Brendan can add. But the -- so we basically -- we have an assumption of approximately 75% of the loans will be forgiven, and 25% will go to term. That seem, I think, pretty typical, but that's our -- the way we think about it.

And as it relates to the profile through the rest of the year, we do think that more will be forgiven in 4Q than 3Q overall. Yes. So maybe a few months ago, we might have thought otherwise, but things have really knocked on as quickly as maybe we thought it was going to go 60 days ago or when this thing all started out. So we do have that layered in. And I'd say that, in general, we were thinking that up to 1/2 of that 75% could come in, in the second half and the rest of it would come in next year. But this is one of those highly volatile things that we're going to keep a close eye on, which is why throughout our materials, we often intend to try to articulate when things look like ex SCB because there's just so much volatility there. So that's what we're thinking about.

[Saul Martinez, UBS Investment Bank]

Okay. No, that's super helpful. So just to clarify then, the 75%, about half and half this year -- second half of this year versus next year, with a little bit -- this year, a little bit more tilted towards the fourth quarter than the third quarter. Is that -- I guess that's a summary of it?

[John Woods, Vice Chairman & CFO]

Yes. Sorry, call it, maybe up to 50%, and that's a good number.

[Saul Martinez, UBS Investment Bank]

Okay. Awesome. Just changing gears a little bit. And maybe I'm overthinking this on the reserving -- in the reserve and your assumptions on provisions. But why is the guidance implying a modest build in the third quarter? And is it -- and some of your peers have talked about your best -- their best guess is sort of that you've now fully reserved based on your -- their assumptions of what the world is going to be and what their lifetime losses is going to be. Is it is it simply that -- are there methodological points that may limit how much you can get ahead of it? Or is there something more fundamental driving that? Or is it just some element of conservatism and you just want to prepare the market or prepare people for the possibility of that if the economy worsens a little bit? Just a little bit more color. And I just want to make sure I understand what you mean by reserve build and what's driving that.

[John Woods, Vice Chairman & CFO]

Yes. I mean I think, as we said in our remarks, we're balancing two overall concepts in CECL. One is if you have your scenario and you're accurately forecasting the performance of your portfolio, then you -- technically, you would only be providing for loan growth. And so on the one hand, you would be releasing, in many respects, reserves. On the other hand, we're cognizant of -- we did this in the first quarter, that should have been true in the first quarter and where everyone built. But then everyone built again in the second quarter, things got worse.

We're cognizant of there's so much uncertainty that we've articulated, including the performance of our portfolio could deviate from our outlook; including our mid to macro scenario in the May baseline for Moody's could be worse when we get to September baseline, as well as our other internal scenarios that we look at and the way that we forecast this; our loan growth trajectory could vary. There's just so many variables that we're trying to balance it by saying, hey, listen, there's -- when you look at the probability-weighted outcomes of all of those scenarios, it's very possible we could have a build. That build could vary and we could have a build outside of a loan, that which was -- would be required for loan growth. And that would -- that growth could vary for all the reasons that we said. And that's really the color around that, reserve build could exist and could be lower than it was this quarter.

[Saul Martinez, UBS Investment Bank]

Yes. Just a final quick one then on the reserve view. I mean it seems -- even without the sales of loan, you kind of moved reserves around where the consumer books, resi, auto, I think, reserve ratios came down and really buttresses are atop CRE and C&I, is that just a function of the consumer books just performing better than you expected and took that as an opportunity to maybe shore up your commercial reserves? How do we read that sort of recalibration there?

[John Woods, Vice Chairman & CFO]

Yes. I mean I think that's right. I mean I think that with all of the forbearance and stimulus that's been out there, I think without improvement -- I mean, consumers and customers have become very flushed with cash. And we've done a pretty granular analysis of our customer base's balance sheets and what their cash levels are and showing up in all of our DDA and everything else. And the credit quality of our back book in consumer has improved at the end of the second quarter as compared with the end of the first quarter. So that has been really an offset to the fact that the macro has actually worsened.

So we have, on one hand, macro worsening, which, all else equal, would have caused the need to increase reserves in consumer, have been more than offset by those factors that I just described. And separately, the commercial customer base has actually -- we're seeing some stress and strain there. So that's actually additive to the worsening -- in a worsening macro environment, and therefore, we had a meaningful increase in commercial builds required as a result of that.

[Operator]

And our next question comes from the line of Peter Winter with Wedbush Securities.

[Peter Winter, Wedbush Securities]

You guys have done a lot of work hedging to protect the margin against the lower short-term rates, but long end of the curve continues to be under a lot of pressure. I'm just wondering what are some things you can do to protect the NIM from a lower-for-longer rate environment. And I heard you on the deposit side, still a lot of opportunities there, but just wondering what else is there.

[John Woods, Vice Chairman & CFO]

Yes. I'll go ahead and just comment on that. I mean I think that our BSO program on the asset side of the house, where we've been optimizing our portfolio investments in the education and merchant space, has been an area that has gone quite well. Even auto these days, in terms of the pricing in auto, has really become much less of a drag and is actually flipping over to be contributing in the NIM space. So we're net -- on net asset side, this has been a multiyear and it's hard and it requires -- it's a multi, multiyear process to really reshape the profile of the asset side of the balance sheet. And so that is clearly a big opportunity. I'd really double down on the point that you made on deposits. I mean when you think about where we were in ZIRP for deposit costs, I mean, we have a clear opportunity to actually lower our deposit costs in the very near-term over the next couple of quarters below where we were in ZIRP, even with the fact that there's a tailwind to the fact that all of our CDs haven't repriced yet. So I'd emphasize that as an opportunity.

And then lastly, I mean, we have a pretty balanced book. I mean our fixed portfolio versus floating is about kind of equal. It's about 50-50. And so we do have -- the NIM is buttressed by the fact that we have that fixed duration in on the loan side, and the fact that we do have hedges that remain in effect actually through the large majority or a large part of 2021. So there's a number of factors that we have in place to try to protect net interest margin, but more broadly, I mean, margins come down in a ZIRP environment. We think we'll moderate it. We could be a good tool to really moderate it, but we're going to be -- nevertheless, we can't really define gravity forever. We're going to be, at the same time, watching us after the tone a little bit to the fee space. And to expenses, we're just -- we've demonstrated a strength in expenses. And so really, we look at it as a PPNR resiliency effort overall going forward.

[Peter Winter, Wedbush Securities]

And just if I could follow-up on loan deferrals. I'm just curious, have you made any changes to the process when a customer comes to extend a deferral? And then secondly, which asset classes are you kind of seeing the most pressure to extend these deferrals?

[Brendan Coughlin, Head, Consumer Banking]

So, yes, I can take that. So we've got a full process around the first 90-day forbearance rolling over to another 90 days. What we try to do is engage with the customer and kind of walk them through the pros and cons from their perspective of extending it. You're going to capitalize interest, you're not paying down your principal, are you doing this for a safety net reason? Or do you really actually need it because you're impaired? And try to coach them through if they're doing it from the perspective of just anxiety or thinking that, that it may be better to reengage and start paying back because you'll be better in the long run, pay less interest.

And we've had some success doing that. But at the end of the day, we've made the decision to be a little bit more liberal with our customers and with the uncertainty in the economy and reopening in question on the speed and timing, if the customer ultimately really wants that safety net of another 90 days to ensure they land on their feet or that they don't get knocked off a little bit, then we're granting that extension of forgiveness. I would say that we've seen about 20% of the forbearance portfolio already rolled off and the majority -- vast majority are reengaging in payments right away. So we're pleased with the early results. It's early innings. The bigger vintages are going to start rolling through here in the mid-summer. So we'll know a lot more in the next 60 days.

[Bruce Van Saun, CEO & Chairman]

Okay. Great. We've got one last time for one last question. Thank you.

[Operator]

And our last question comes from the line of Brian Klock with Keefe, Bruyette, & Woods.

[Brian Klock, KBW]

Just one real quick one on the expense. I just wanted to ask John on -- you guys have done a really good job on controlling expenses. And definitely in this environment, it's pretty difficult, so good work on that. And I just wanted to think about the math. For the full year, you're saying it could be up modestly year-over-year. So it seems like when you put in your third quarter guidance that you're implying some level of operating expenses that's probably near -- that's closer to $900 million, and that's pretty low relative to where you've been. So does the math makes sense? And is that the sort of run rate to set up for next year for the fourth quarter?

[John Woods, Vice Chairman & CFO]

Yes. I mean I think you've got to look at -- we said that we'd be up modestly in 2019 versus '20 -- I'm sorry, in 2020 versus 2019. And there's investments that are being made in the platform. There is -- 2020 is a big mortgage-origination year, so there are significant expenses, of course, that you have to incur when you're generating the revenue levels on the fee side that we are. So we're keeping an eye on that, but it's offset by the investments that we're making going forward.

In 3Q, we said, again, we did reference the mortgage expense number, but as I mentioned before, we're taking -- we're looking at all of this in the TOP 6 arena. 2021 is a big year for TOP 6 in terms of what we expect to accomplish there, and we're looking for ways to actually add initiatives to that program and to possibly offer ways to continue to manage those expenses going forward.

[Bruce Van Saun, CEO & Chairman]

I think that exhausts all the questions. I do want to thank everybody for dialing in today. We always appreciate your interest and support. Have a good day, and, everybody, stay well. Thank you.

[Operator]

Thank you. Ladies and gentlemen, that does conclude your conference for today. Thank you very much for your participation. You may now disconnect.