Welcome to the OneMain Financial Second Quarter 2020 Earnings Conference Call and Webcast. Hosting the call today from OneMain is Kathryn Miller, Head of Investor Relations. Today's call is being recorded. [Operator Instructions] It is now my pleasure to turn the floor over to Kathryn Miller. You may begin.

Thank you, Maria. Good morning and thank you for joining us. Let me begin by directing you to pages 2 and 3 of the second quarter 2020 investor presentation, which contain important disclosures concerning forward-looking statements and the use of non-GAAP measures. The presentation can be found in the Investor Relations section of our website. Our discussion today will contain certain forward-looking statements reflecting management's current beliefs about the company's future financial performance and business prospects, and these forward-looking statements are subject to inherent risks and uncertainties and speak only as of today. Factors that could cause actual results to differ materially from these forward-looking statements are set forth in our earnings press release and include the effects of the COVID-19 pandemic on our business, our customers and the economy in general. We caution you not to place undue reliance on forward-looking statements. If you may be listening to this via replay at some point after today, we remind you that the remarks made herein are as of today, July 28, and have not been updated subsequent to this call. Our call this morning will include formal remarks from Doug Shulman, our President and CEO, and Micah Conrad, our Chief Financial Officer. After the conclusion of our formal remarks, we will conduct a Q&A session. So, now let me turn the call over to Doug.

Thanks, Kathryn. And good morning, everyone. This remains a challenging time for everyone. The COVID-19 pandemic continues to impact every aspect of our lives. I would like to thank the OneMain team members who continue to provide the highest quality service to our customers and support to each other through these unprecedented times. OneMain remains focused on the wellbeing of our customers, team members and communities. Our highly flexible omnichannel model is well-positioned to adapt to current market and public health conditions and ensures we can continue to serve our customers. Supporting our customers through good times and bad is a hallmark of OneMain and how we have gained the trust of millions of customers. As we've said before, OneMain has several fundamental strengths. And these strengths have been apparent through this crisis and are evident in our second quarter performance. They include our deep experience and proprietary data, supporting best-in-class underwriting and analytics, a hybrid operating model that optimizes the customer experience and performance, our strong conservative balance sheet and deep capital markets access and our superior return profile and consistent capital generation. We continue to be guided by our key operating principles, strong and stable credit performance, disciplined originations and a conservative balance sheet. Given the continued uncertainty in the economic climate, we are being conservative in how we manage the business. In early March, we implemented the downturn planning playbook that we had developed in 2019. We took quick action and tightened our credit box, underwriting only loans that met our return hurdles in a 2008/2009 recession scenario, which included cutting out the higher risk unsecured lending, increasing income verification requirements, and adjusting collateral and net disposable income requirements. Since then, we've refined our underwriting even more by adding stress factors to certain high risk industries, such…

Thanks, Doug. And good morning, everyone. Let's move right into our second quarter results. We earned $89 million of net income or $0.66 per diluted share. On an adjusted C&I basis, we earned $107 million or $0.80 per diluted share. Originations for the quarter were $2 billion, down 47% from the second quarter of last year. As Doug mentioned, we took quick action in early March, tightening our credit box and safeguarding our portfolio, which in combination with lower demand, led to the decline in originations for the second quarter. This tightening also led to a 57% mix of secured originations compared to 55% in last year's second quarter. Customer demand and loan originations have steadily improved as consumer spending and employment have improved throughout the quarter. June originations were down 26% from a year ago as compared to our April originations, which were down 63% year-over-year. Our ending net receivables were $17.7 billion for the quarter, down $551 million sequentially, but still $716 million or 4% higher than the second quarter of 2019. Assuming current trends, we expect the rate of receivables liquidation to decelerate in the near term relative to the $551 million liquidation we experienced in Q2. Interest income was $1.1 billion in the second quarter, up 7% from last year, primarily reflecting 8% higher average receivables compared to the prior-year period. Yield was 8 basis points lower than last year's second quarter, reflecting the impact of increased borrower assistance. However, yield was up 2 basis points sequentially as the impact of borrower assistance was more than offset by the sequential improvement in late stage delinquency. Interest expense for the quarter was $266 million, up about 14% versus the prior year, reflecting a $4 billion increase in average debt balances. A third of this increase supported the…

Thanks, Micah. Let me close by reiterating what we've been saying for quite some time. Our business model and the value we provide to our customers gives us real competitive advantage in the consumer lending space. And while there are uncertainties that lie ahead, we believe that we are very well positioned for success. The core strengths of our business and the strategic initiatives we are driving are enabling us to improve the ways in which we serve and support our customers while optimizing our financial performance for long-term value creation. With that, let me thank all of you for joining us, and I'll turn the call over to the operator for questions.

Thank you. [Operator Instructions]. Our first question is coming from Michael Kaye of Wells Fargo.

Good morning. Does that lower 2020 net losses represent just a push out of the timing of losses until 2021? Or are we seeing a real improvement in the credit loss trajectory from what you contemplated back during the Q1 earnings call?

Yeah. So, Michael, good morning. This is Micah. Obviously, credit is performing well. I would say better than expected, frankly, with our 30 to 89 day delinquency at 1.63 for the quarter. Doug mentioned that was the lowest quarter we've had in the history of the combined company. So, really tough to say beyond the end of this year. I think we feel really comfortable just because of where the roll rates are and where our early delinquency is performing, calling out the 5.8% to 6.0% this year. There's a lot still to be known here. We're in the early stages. And I think it's too early to say what's going to happen in 2021. But we feel comfortable at this point where we are for this year.

Okay. I wanted to see what you're observing in the competitive environment right now. Is there an opportunity for you to move up perhaps like higher FICO customers that you can make attractive risk reward returns?

Yeah. Hey, Michael. How are you? It's Doug. I think a number of competitors have pulled back. I think most competitors have tightened their credit box like we have. We have the advantage over some of the competitors, which is we built a more liquidity and a longer liquidity runway. So, there are a number of competitors who are originating less for the same reasons we are because we tightened our credit box and demand is down. But the third reason is there's some that just don't have the money or the cash to make loans. We have seen in what I would call the lower end of prime or the higher end of non-prime, kind of the 675 to 725 FICO scores, competitors take their pricing up. Our pricing has always been relatively competitive there. And so, we are picking up some share in that kind of range, but the higher quality credit where some competitors may be pulling back for a number of reasons. And we're definitely focused there because it's good, high quality credit. So, those are some of the dynamics.

Thanks very much.

Our next question comes from the line of Eric Wasserstrom from UBS.

Great. Thanks. Good morning.

Good morning, Eric.

Just to follow up on the prior question around the cadence of losses. It seems as if many in the industry are basically suggesting one of two loss curves that are really more like step functions. One is an expectation of a fairly benign third quarter, but a more significant step up in the fourth as stimulus abates and deferrals and borrower assistance – the benefits of that diminish. And then, the alternative model is more of a peaking at some point next year as those same effects diminish and loss trends have more converged on unemployment. And I'm just wondering if you have a view on where OMS might fit in among those different options.

Right. So, it's a good question. Thanks, Eric. And good morning. So, let me just step back for a moment on losses. So, in order for us to see losses, we first have to see delinquency. Right? And we have not seen delinquency rise in our portfolio. I mentioned in Michael's question that where we were as far as the second quarter early stage delinquency, we have 180 day charge-off policy and we're about 180 days from the end of the year. And so, really, what needs to happen to move the needle on 2020 losses is either a change in early delinquency. And that's, I would say, behind us at this point just from a flow rate perspective. And then what happens with more of our late stage delinquency and how many of those loans roll to loss. So, again, in terms of the year, when we're talking about losses, we really feel like we have these losses in the buckets today and it's not going to move the needle that much on the full year loss number. Now, with respect to 2021 and your question around the step function in the third or fourth quarter, certainly, that will be a function of job creation and where we are with government support, which we're not sure yet where that's going to land from an extension point of view, but I think people are referring to levels of delinquency at that point, which will then influence your 2021 losses. I would say, our CECL reserving, we've assumed just that. So, we start with our reserving with a baseline loss forecast and then we implement stress on the portfolio. We've assumed in that reserving that there is no extension of government stimulus, as I mentioned in the prepared remarks, and we assume in that reserving that the delinquency rates begin to deteriorate pretty immediately. That's a reserving point of view. We haven't seen that yet in in our results, but that will be more of an influence on 2021 than it will on 2020, if that's helpful.

Yeah, very helpful. Thank you. And if I may, just follow up with a question on the borrower assistance programs. Again, Micah, could you potentially give us any granularity on, of the 85% that are no longer in the program, how many flowed to current, how many moved back into delinquency and how many potentially flowed to loss?

Yeah, sure. As you know with our programs, they're individualized and tailored, really come with a combination of single pay deferments and also shorter-term modifications to help customers to a little bit longer term of stress. I would make two important points here on borrower assistance. One is that this isn't new for us. We average borrower assistance enrollments of about 2% per month normally and we've been using these programs for a long time. The second point is that, in addition to them being tailored, we almost always include a partial payment from customers. And so, we talked about those 85% of customers who have rolled off, I would say, as evidenced in our delinquency rates for the month, those are performing as we expect. We're seeing very, very strong payment rates, as you can see, in our earnings presentation in June. So, that indicates that a lot of those customers coming off of borrower assistance are paying. I will tell you those trends remained fairly consistent in the month of July from both a payments and delinquency standpoint. So, it's too early to tell how those loans and borrower assistance will ultimately perform. But given our underlying delinquency and payment trends in the portfolio, we feel very good about where it is right now.

Thanks very much.

Thank you.

Our next question comes from the line of Mark Hammond of Bank of America.

Thanks. Hi, Doug, Micah and Kathryn. With public equity valuations of consumer credit companies materially lower this year and your significant liquidity position, I was wondering if I could get your take on what you're seeing as far as inorganic growth opportunities in the market right now or if you think they all have yet to come and will come in the future?

Yeah. Thanks, Mark. We've been innovating in our platform all from a customer perspective, which is what else can we do for our customer, how do we attract more customers, how do we make sure the value we provide to customers continues to grow. And so, our lens for either product innovation or evolution or different channels, whether it's digital channel, phone channels, partnership channels, it's kind of the areas that we've been exploring and I've talked before about some of those. There are a number of companies that need capital and put themselves up for sale. Our M&A team is pretty active in conversations. What I would say is until things shake out a little bit more, we're not in a big rush. If we see the right opportunity, we'll potentially be able to pick up either a platform or a product company. I've mentioned before publicly that our bias would be more to smaller tuck-in kind of situations. And so, all I would say, there is still a lot of uncertainty out there. There is a lot of activity and we're open to discussions, but there's nothing imminent at this time.

Thanks, Doug. And just a follow-up. Besides the platform, I was thinking years ago about the SpringCastle portfolio where it was just a good opportunity at a good price. Have any of those come across OneMain's desk yet?

Yeah. Look, there's some portfolios that are out in the market. As we've said, we think there's still a lot of uncertainty in the market. We do think a number of factors including the government stimulus are making the correlation between delinquencies and losses and unemployment not equalized yet. And we'll have to see where that goes. So, I think you'd have to be very careful about buying portfolios in this environment. With that said, when there are portfolios, we've got a servicing capabilities technology to do it expertise and we know how to bring on portfolio. So, there are some out there. We'll keep looking at them, but we also need to take into account that right now there's still a lot of uncertainty in what the future looks like.

Thanks, Doug, Micah and Kathryn. That's all.

Thank you.

Our next question comes from the line of David Scharf of JMP Securities.

Hi, good morning. And thanks for thanks for taking my questions.

Morning.

Doug, the first, actually just to provide some context, in the course of discussing sort of tightening the credit box and underwriting, you had specifically called out certain verticals like travel and leisure, entertainment and dining. I'm just curious, is there a rough percentage of the number of borrowers that are employed in those verticals right now or maybe where it's been at a peak and where it is now?

We don't have that information. We know when people are coming in who their employer is, so we can identify a vertical. I think, nationally, the number of borrowers that are out there in those verticals, I don't have.

Got it, got it. Understood. And then, shifting gears on the origination front and the remote closing, I'm curious, of the roughly a third of volume in the quarter in which borrower didn't actually touch the branch physically, were any of those new to the company? Are those primarily servicing kind of existing customers that were kind of reupping their existing loans? And if that's the case, if you anticipate over the course of the next year kind of expanding that service to new borrowers to OneMain.

Yeah. No, thanks for that question. First, I want to be clear. We give customers the choice and our goal is to make it easy for customers to do business with us, while at the same time we make sure that customers who did business with us meet all of our requirements to extend credit to them. The remote closing is heavily weighted towards existing customers because it's quite easy, we have a lot of information on them, they already have an account with us, it's very easy for them to close remotely. There are a good number of new customers. And just as a reminder, when we bring in either a new or present customer, they go through the same process in a branch with a detailed discussion with a team member, a budgeting exercise that determines ability to pay, income verification, and we built the technology to upload and verify documents. We've got video assisted, so we can have a video call with them. And it's weighted more at this point towards unsecured customers because to secure – to have security on the automobile, you have to get access to the title. And so, we do have a procedure to get the title, but if somebody comes into a branch, they can get funds the same day because we see the title, we inspect the car and cash is disbursed. If they're closing remotely, they might have to wait a day, either to overnight the title, drive the title through, we can upload pictures, et cetera. So, what I would say is, this is not new for us. This is just something that the unique circumstances of the pandemic gave us an opportunity to be available to clients and customers in a unique way. And so, we had built all of this capability before. I do think it will be an important part of the company going forward, which is giving customers an option of channels to interact with us of the channel of their choice. And so, it's worked out well and we're happy we had this all set up, so we could serve customers through this difficult time.

Got it. Understood. Very helpful. Thank you.

Our next question comes from the line of Vincent Caintic of Stephens.

Hey, thanks. Good morning. Just wanted to get your thoughts on the second round of government stimulus. So, the Republican proposal came out this morning or late last night and you're talking about a second round of $1,200 checks and maybe some additional benefits. Just kind of wondering maybe scenario analysis, your thought on impact from any of these proposals.

Yes, thanks, Vincent. Having spent some time in Washington, I don't hazard to guess what two houses of congress and an administration will end up landing on. So, I don't have any great insights on where to land. I think, obviously, the government stimulus gave broadly the American consumer more spending power. Most consumers and a lot of our customers we've talked to are prioritizing paying current bills above large purchases. I think more stimulus is better for consumers meeting their financial obligations and we're just going to have to see what happens over the next hopefully week or so.

Thanks. Maybe following up on that, so we can see the benefits on credit. Any thoughts on – because it seems like on the originations and demand side that your demand is actually doing well even with customers having saved more. So, any thoughts on how you're thinking about the demand curve going forward with the potential stimulus or kind of going into 2021?

Yeah. Look, if you look at what happened with demand, I think late March and into April, the whole country was moving into lockdown. And I think people were in kind of the shock of the beginning of the crisis. And people were just restricting all of their activities, moving around, spending, buying, anything like that. In May, I think as people got used to, 'oh, this is the new normal and it might last for a while,' as well as different states started opening up their stay at home orders and people started moving around a little bit, people got back to – or started to move in the direction of more normal buying habits and borrowing habits. Generally, when customers are feeling good about their financial situation, they're willing to take out a loan to do something that – either a home repair – obviously, there's some emergency things, whether it's medical or a water heater breaks down. I think people are very interested in debt consolidation and that remains part of what's happening. And so, I think stimulus will have an effect. I think the bigger effect is probably on delinquencies and losses than on demand. But, obviously, more money moving into the economy makes people feel better, there's more spending and that will probably have some impact on originations. I think on originations, look, we're now in a recession or technically we might not be in one, but we feel like we're in a very murky economic climate that could last a while. We really don't manage to growth. We set our credit box, we set it based on a 20% ROE, return, we tightened our credit box now, but we'll still do our marketing and make sure once somebody is interested in the loan, we provide a great customer experience, so they can move through the company, talk with someone, see their options, and if they qualify, we'll give them a loan. So, it's very hard to say. Obviously, originations have picked up. That seems to be holding in July. I think a lot of this is what happens with the pandemic and consumer sentiment and mindsets going forward.

Okay. Thanks, Doug. Very helpful.

Our next question comes from the line of Kevin Barker of Piper Sandler.

Good morning. Could you just help us understand the full impact that we're seeing from the deferral of tax refunds that typically would come through in March and April, and how that's impacting just the potential paydowns on your loan book and your origination volumes that we're seeing today?

Hey, Kevin. It's Micah. Thanks for the question. Really hard to say where funds come from, right? Certainly, our customers, we do see normally in March an increase in payments associated with tax refunds. That seemed to be a little disrupted this year with what we saw at the end of March with respect to delinquency. You might remember we saw delinquency rise as people were getting dislocated. So, I think that whole timing dynamic of deferrals on tax refunds certainly had an impact. If you just look at the page we put in the earnings presentation, you'll see how strong our payments have been. On average, we will run about 4.6% of the portfolio in any given month. We were at about 4.5% in March and that ticked up to about 4.8% in June. I'll remind you of that. Let's just round it to 5%. Roughly 2% is interest. 2% is regular payments and a percentage point of prepays is the way that breaks down. And I think we've seen strength in payments. We are comfortable with where they are. Really hard to say where those funds are coming from, whether through tax refunds or just government support.

That's helpful. And then, I appreciate the month by month origination trends that you put out there. Could you give us some early indications on how July is trending, given the positive results you start to show in June?

It's similar to June.

Okay. All right. Thank you for taking the question.

Thanks, Kevin.

Thanks, Kevin.

Our next question comes from the line of John Hecht of Jefferies.

Thanks very much. And good morning, guys. Congratulations on a good quarter. Most of my questions have been asked. I guess I'm a little curious. OneMain, I bet you guys have been investing in your digital infrastructure for as long as I can remember. Doug, I think you were focused on enhancing the consumer experience by deploying some capabilities that enhance convenience within the branch when you first got there. Obviously, there's a transition. We're accelerating more digital interactiveness. I'm wondering, is there any kind of – I think you've identified in the near term that maybe change your priorities about how you want to develop them and deploy them on that side of the business just to meet the changing kind of consumer behavior?

Yeah, it's a good question. Look, in March, when the stay at home orders happened and a lot of our corporate teams, technology, marketing, analytics, strategy, finance teams, et cetera, started working from home and we anticipated that we were moving into a different stage of the cycle, we did a full reprioritization in the company. And as Micah said, we took out some expenses that we didn't need as originations went down and there was going to be less business coming through the pipe. In the short run, we wanted to make sure we were prudent with expenses. But we also looked and said, which of the investments we've been making do we want to accelerate? Because we want to make sure we keep investing in the long term franchise, that we're one of the stronger players in the space, and we're going to be able to keep originating loans through the cycle and obviously accelerating out of the cycle. For sure, our investments in digital, technology, analytics were top of mind. And so, if you think about the digital experience, people apply usually online. So, what does that application look like? What do our web properties look like? What is our mobile device – and many customers are interacting with us on their mobile device. Some are using a browser on their mobile device. Some are using the app. And all of those things we just mentioned, we're making sure that all along the stages of the experience with OneMain, so it's the application, it's the funding, it's questions they have, it's servicing they do. We've built out a set of robust consumer financial education tools that you can use, you can access it, budgeting, planning for a vacation, how to operate in a downturn, how to manage…

Great, that's helpful and very interesting. And I guess, similarly, you guys are becoming more precise on your underwriting capabilities. And people are using big data and AI and things like that on the front end. Maybe any update on that or anything that we should be looking forward to in the coming months or any just commentary about what types of data sets you're using that have enhanced your ability to operate in this kind of environment?

Yeah. Look, one is because this recession is induced by a public health emergency and a pandemic versus the last one, which was a capital markets, financial markets, housing induced recession, this one's more physical, we think there's going to be some asymmetric effects on the consumer. And so, I mentioned some of them. We actually pivoted and didn't used to do underwriting by industry. We quickly built the technology, got some external data sets. We're now looking at a lot of data sets of what's happening by industry. There also could be some opportunities for industries that aren't hard hit or industries that as we move through the stages of the pandemic actually are adding employment where we can underwrite in a differentiated way. Right now, we're using it defensively. We could potentially use it offensively. Almost all of our underwriting is supported by AI and machine learning where the algorithms learn from each other and in real time can kind of look for patterns and see these. We ran in – like starting in 2015, we used to run classical regression testing, algorithms for underwriting. Over time, we ran kind of champion challenger models where we saw both. We saw some boost and we had some pages in Investor Day of the different kinds of AI that we use for underwriting. And our modeling teams are always looking for new opportunities. We actually are partnering with a couple interesting more fintech type players who specialize in finding unique correlations of either macro or consumer factors for underwriting. And so, we've got a large data set, we've got good underwriting, we always have tests going with third parties to see how we can enhance that. And so, I'd say, over time, we've seen a lot of innovations. I'm not sure there's any specific pandemic-induced ones except for the pandemic because there could be an asymmetric recovery by geography and industry, having those kinds of tools to use, so you can real time update your underwriting will be important.

Really appreciate Thanks very much, guys.

Thank you.

Our next question comes from line of Eric Hagen of KBW.

Thanks for taking my question. Good morning. How should we think about the tweaks to the underwriting criteria that you mentioned in your prepared remarks and its impact on the portfolio yield going forward? And then, additionally, you guys have noted a lot of success with renewing borrowers as their loans roll over. So, with regard to those same underwriting tweaks, how meaningful do you think those tweaks might be for some of that cohort which has renewed with you in the past to qualify for funding based on the new criteria?

Yeah. Look, like I said, we tightened our underwriting. We did it with kind of blunt instrument. Let's no regrets move, tighten the box in March. Since then, we've taken more of a scalpel and refined it. One of the things I mentioned is taking out higher risk, unsecured, is one of the first moves we made because it has higher loss content. And so, under a more 2008/2009 type stress scenario or even more than that, it won't meet our risk return hurdles. As a result, I think you could see more secured, as Micah told you. You saw more secured because more people only qualified for a secured loan. Secured loans have lower yield. So, over time, you could see a little bit lower yield in the portfolio because of that. With that said, we also think we'll be able to pick up some unsecured business with higher FICO scores that – even though we don't underwrite to FICO scores or better credit quality. So, I think it's going to be hard to say how that mix will sort itself out. Micah, you might want to add?

Yeah. No, I would comment on the yield. I think, certainly, we saw 57% secured in the quarter just because of some of that credit tightening. It will impact yield over time because it's a lower ATR product, but it's going to take a while just simply for that to find its way through our significantly-sized portfolio. That would be one. And with respect to the renewal question, we go through another set of underwriting with our renewals. So, once a loan comes on, we'll see how it performs in the vernacular of on-us performance, which is how do we see payments, has the customer been delinquent, the ultimate outcome of whether we see a change in renewals because of the tightened criteria is still, I think, yet to be seen.

Got it. Thanks for that. And then, it's nice to see the capital distribution this quarter. I guess it's in the third quarter. And I hear you that through year-end, the earnings will likely be strong. But what were the variables that led you to decide to distribute capital now versus keeping a little bit more of a cushion heading into 2021, which I think you acknowledge still had its fair share of uncertainties?

Yeah. No, thanks for that. Look, we are very mindful of the uncertainty in the economy. But we've taken a number of actions to position the company defensively. We've tightened credit. We increased reserves. We continue to operate at the lower end of the leverage range. We trimmed expenses. And we performed further stress testing, assuming very severe downturns and things turning. We pushed it very hard. And so, we feel super comfortable with our capital position and the cushion we have. We've also taken some good offensive moves with the business, investing in omnichannel and digital I talked about, some product innovation and picking up some good, higher credit quality customers. And so, I think the decision that we took with our board is that we should stay disciplined in managing the business. And given our results, given the cushion we have, given the defensive posture, and given our capital profile, we want to stick with our capital return strategy.

Got it. And just to clarify on expenses, when you say lower than last year, is that on a gross basis or on a percentage of receivables basis?

Gross expense.

Okay, thank you, guys, for the comments.

Thanks, Eric.

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

Great, thanks. Most of my questions have been asked and answered. But we have noted that you did some pretty consistent mail volumes even during the second quarter while the rest of the industry pretty much went close to zero. And you've talked a little bit about potential for some share gains, but maybe just talk a little bit about what could get you back to something kind of even closer to kind of normal levels of origination. Obviously, you talked a little bit about the stimulus potential being a little bit of an impediment in the near term. But as you kind of look out at the back half of the year, any further thoughts?

Moshe, thanks. Good morning. Let me start with the mail volume and then Doug may want to add some thoughts around what we might see in terms of future originations. But we're familiar with the data you're looking at. I would tell you that the data is not entirely accurate. I would also say, as OneMain does not have liquidity challenges that some others on that list might have, so that certainly could influence the appetite for doing mail. We run a diversified marketing strategy and we have direct mail, but we also have many affiliates and digital channels, as well as our own organic acquisition. And all of our mail, as we talked about with our risk-adjusted returns and our underwriting framework, we send our mail based on ROE thresholds. And that remains a very effective channel for us and we expect to continue to – we expect it to continue to perform as expected.

Yeah. And hey, Moshe, to where do originations go, I don't want to be a broken record, but I will be one that we really don't manage to a growth number. We manage to a – Moshe, I don't know if you want to mute your line, there's a lot of feedback. We manage to book every loan that meets our risk return criteria. With that said, as Micah mentioned, we're continuing with mail, but we've adjusted it. A bunch of competitors have moved out of digital marketing and some of the paid digital and purchasing of keywords, et cetera, like personal loan and other things, the prices have gone down significantly during this pandemic and so we've actually seen an increase in the ROE on our digital marketing spend. And so, we're going to continue to optimize our marketing, then optimize – once somebody gets interested in us – having a great experience with us, and then if they meet our credit criteria, we'll book it. I think given our tighten credit box and given the uncertainty in the economy and consumers being pretty cautious, we don't expect to get up to last year's levels in the near future. But as you've seen, the trend is going in the right direction. And we're going to book the loans that meet our risk return criteria.

Great. Thanks, Doug. And, Micah, just a quick follow-up. I know you mentioned that third quarter interest expense should be kind of similar. Any things that you're doing from a funding standpoint that could benefit in terms of future cost of funds?

Yeah. As you know, Moshe, this is a long-term strategy for us with our balance sheet, our emphasis on liquidity. I think the big driver of what we expect with third quarter and fourth quarter is just the redemption of that 8.25% 2020 bond that will be paid down as of July 29. So, that's really the influencing factor. We're going to continue to run the business and use opportunities to issue where we can.

Okay, thank you.

Thanks.

And our next question comes from one of Bill Ryan of Compass Point.

Good morning. And thanks for working me here at the end of the call. Question on involuntary unemployment insurance. If you can maybe give us an idea of maybe what percentage of the portfolio that carries it, how it's activated, and then once it's activated, how it sort of impacts the various line items in the P&L. I just noticed, in going through the numbers, a couple of line items were quite a bit higher than expectations on the revenue line. So, I was just kind of curious how all that kind of came together. Thanks.

Sure. Yeah, let me try to unpack that a little bit and happy to do a follow-up with you if required for a little bit more detail. But on our involuntary unemployment insurance, we've talked about the portfolio having about 25% penetration overall of that product. The way it arises in terms of an expense being incurred is we have either an expectation for our reserving that IUI claims are going to be made or a customer actually makes one. So, very much a reserving process that influences our expenses. If one of our borrowers or customers were to lose their job, in that unfortunate situation, the IUI covers the payment on their loan for up to 12 months. It requires very similar requirements to state unemployment. You have to be claiming state unemployment and you have to be proving and demonstrating that you're actively looking for work. So, very similar in that regard. We don't disclose the number of claims that we're seeing. But I can tell you, as I mentioned in the prepared remarks, we saw claim levels peak in April and they are down about 62% in June, which is pretty consistent with what I think broadly we're seeing in initial US jobless claims. So, we're seeing some positive trends in final claims payments, which indicate a return to work for many customers. All that being said, we're still in the early stages. In terms of how it shows up in the income statement, you're largely going to see it through the policyholder benefits and claims line, which we mentioned was up significantly year-over-year and also sequentially, almost entirely due to that, those IUI expected claims that we have in that line item.

Thank you.

Thanks for the question.

And thank you, ladies and gentlemen. This does conclude today's OneMain Financial second quarter 2020 earnings conference call. You may disconnect your lines at this time and have a wonderful day.