CURO Group Holdings Corp (CURO) 2018 Q2 法說會逐字稿

Good morning, and welcome to the CURO Group Holdings second quarter 2018 conference call. All participants will be in listen-only mode. (Operator Instructions) After today's presentation, there will be an opportunity to ask questions. (Operator Instructions) Please note, this event is being recorded.

I would now like to turn the conference over to Gar Jackson, Investor Relations for CURO. Please go ahead.

Thank you, and good morning, everyone. CURO released results for the second quarter 2018 yesterday evening after market closed. You may obtain a copy of our earnings release from the Investor Relations section of our website at ir.curo.com.

With me on today's call are CURO's President and Chief Executive Officer, Don Gayhardt; Chief Operating Officer, Bill Baker; Chief Financial Officer, Roger Dean; And Chief Accounting Officer, Dave Strano. This call is being webcast and will be archived on the Investor Relations section of our website.

Before I turn the call over to Don, I would like to note that today's discussion will contain forward-looking statements based on the business environment as we currently see it as of today's, July 31, 2018, and, as such, does include certain risks and uncertainties. These statements relate to our view of our customers' financial health, our expectations for loan demand, our expectations regarding number and timing of opening new branches in Canada and the deployment of our open-end product in Canada, our expectations related to the impact of the 2017 Tax Act, and our 2018 full year financial and effective tax rate outlook.

Please refer to our press release and our SEC filings for more information on the specific risk factors that could cause our actual results to differ materially from the projections described in today's discussion. Any forward-looking statements that we make on this call are based on assumptions as of today, and we undertake no obligation to update these statements as a result of new information or future events.

In addition to U.S. GAAP reporting, we report certain financial measures that do not conform to generally accepted accounting principles. We believe these non-GAAP measures enhance the understanding of our performance. Reconciliations between these GAAP and non-GAAP measures are included in the tables found in our earnings release. As noted in our earnings release, we have posted supplemental financial information on the IR portion of our website.

With that, I would like to turn the call over to Don.

Great. Thanks, Gar. Good morning, everybody, and thanks for joining us today. This call will follow pretty much the same drill as in the past. I'll offer some high-level thoughts on the quarter, a few strategy notes, and a few brief comments on the regulatory environment. Roger will then give you much more detail on the numbers, and then we'll take some questions.

I would characterize our second quarter in two ways. First, from an operational standpoint, we had an excellent quarter. We made great progress on our new loan products, our MetaBank relationship and, probably most importantly, the early stage of a very successful introduction transition of a big part of our Ontario lending business of the legacy single-pay loan product to a line of credit that's been very well-received by our customers.

We'll unpack the Ontario transition in some detail later on, but it's a big undertaking that's going very well and is running well ahead of schedule. We did all this while maintaining our credit and other financial disciplines. These are big projects that require many people from many departments to work together, and we're incredibly proud of all of our CURO team members for giving this a huge effort this quarter.

From a bottom line perspective, the quarter was very good, but the provision build associated with the second quarter asset growth increased advertising expenses for our newer brands and the expenses related to the ongoing affordability settlements in the U.K. combined to keep earnings from coming in at a level that we would characterize as excellent. But the good news is our U.S. business is extremely strong and performing in a way that more than makes up for any shortfall from our international operations.

We are affirming our guidance today, and Roger will go into that in more detail in his prepared remarks, but before he does, I'll just make a few comments on that. As I said, we're very pleased with the performance of our U.S. business, and we have a very high degree of confidence in achieving our guidance objections. And, obviously, today we're about 60% of the way through the year, so that certainly helps. And our confidence in our core business and products is very high. It's probably just that the Canadian product transition and the U.K. affordability issue are both too fluid for us to completely discount a bit of downside risk on both those fronts.

So looking at our current forecast, we expect our Canadian and U.K. operations to fall short of our operating earnings plan for the full year 2018 in the range of $10 million -- that's U.S. $10 million. So sitting here today, we do think, as I just mentioned, the U.S. business will be able to make -- to more than make up for these projected international shortfalls, and we believe that there's a good likelihood that we'll come out ahead on our internal forecasts and our published guidance.

Turning to some quarter and year-to-date highlights, revenue was $249 million, which is 14.8% percent above our year-ago quarter. For the 6 months, total revenue came in at $510.7 million, which is 15.7% ahead of prior year's 6 months total, so still strong growth and running a bit above our longer-term growth goals.

Adjusted EBITDA for the quarter was $48.6 million, or 8.6% behind last year. For the 6 months, adjusted EBITDA was $123.8 million, which is 1.6% ahead of prior year. And Roger will go through this in more detail, but the loan loss provision, based on strong asset growth and the ad spend investment in newer brands, offset the very good revenue growth. The provision for the quarter is also comping against our second quarter 2017, when we released some of the over-provisioning from the first quarter of 2017, when we changed reserving methodology for non-single-pay loans.

Now, we realize that there's a lot of elevator analysis here to keep up with, but the underlying credit trends, the fundamental credit trends, are still very favorable, and both Roger and I will have some more detail on that point.

Adjusted net income was $18.1 million, just slightly off of last year, but still up an impressive 18.9% for the 6 months versus the prior year. Gross combined loans finished the quarter at $513.8 million, or 24.6% greater than the balances at June 30, 2017, and $70 million, or 15% higher than the sequential March 31, 2018, balances.

As expected, single-pay balances fell year-over-year, and we had low double-digit growth in our secured installment and CSO portfolio. However, our combined installment and line of credit portfolios -- we call the Multi-Pay products -- grew by $87.6 million, or 47.9% year-over-year, and by $47.5 million, or 21.3%, in just the last quarter.

The rest of my remarks will focus on 5 topics -- Canada, the U.K., Meta, credit, and Ohio and the broader regulatory and legislative climate.

First, in Canada, as I mentioned earlier, ongoing changes in the provincial regulation of single-pay lending in Canada, coupled with our growing competencies in marketing, underwriting and servicing line of credit products, brought us to the decision to accelerate the transition of our product offerings, particularly in Ontario, which accounts for about 2/3 of our Canadian revenue.

Although we anticipated introducing a line of credit product in our Ontario locations during 2019, early test results we initiated in February of this year were incredibly favorable in terms of acquisition costs, credit performance, take-up rates, line utilization, and really the whole deal, so we moved up our -- we simply moved up our conversions schedule.

So what does that mean? It means higher earning asset balances in our line of credit portfolio in Canada than previously forecasted, but lower yields and lower revenue while the book builds and higher provision when larger balance dollars are originated.

We started the quarter with just over $53 million in non-single-pay -- these are all U.S. dollars, by the way. We started the quarter with just over $53 million in non-single-pay balances, and we ended the quarter with $74.7 million, and having booked the kind of larger marketing and convergence plan in late June, the balances today sit at over $115 million, so great growth, all driven by tremendous customer communication and service by our store and call center teams in Canada.

As I mentioned, short-term, this approach diluted earnings for the quarter. There's really no question we could have had better earnings had we taken a more incremental approach, and it will impact earnings for the full year versus our plan, but long-term, if the portfolio continues to build through the remainder of the year, our exit rate will be very high, and we'll have a much more diverse and stronger business in Canada that will drive substantial operating earnings gains in Canada in 2019.

I'll talk about the U.K. for a second. Like our Canadian team, we're really proud of the second quarter that our team in the U.K. produced -- loan balances up to 31% year-over-year; 29,000 new customers, a 45.3% increase year-over-year; stable credit; and good control of operating costs. All on the back of an improved customer acquisition and application and approval journey that we think puts us ahead of our U.K. competitors and positions us for strong -- continued strong growth.

Unfortunately, as you can see in the segment results in the release, we still posted a $2.5 million operating loss for the quarter for a business unit that we had expected to just about break even for the quarter. This is partly due to the provision build as earning assets grew during the quarter, but the biggest variance is the expenses related to the ongoing and somewhat unpredictable issue of paying redress to former customers claiming that we had not adequately assessed their ability to afford to borrow from us.

Now, I'm leery providing any more we like to say bulletin board material for anyone in the U.K., so if you'd like to know my true feelings about this issue, I'll refer you to our April call transcript.

I don't want to get ahead of myself and say that we don't have any good news to report, but we are cautiously optimistic, based on some recent dialogue with regulators, that they are beginning to grasp the scope of the issue for U.K. lenders, and this is an issue that affects everybody in the small-dollar credit space in the U.K. And also, they're beginning to understand the fact that this process yields scant benefit to consumers.

I hope in subsequent quarters that the financial impact of this process will decline and the operating improvements that our U.K. team has engineered will finally and meaningfully flow to the bottom line.

I have some quick thoughts on Meta, where we continue to make very good progress with them and look forward to getting our VERGE CREDIT, powered by MetaBank, product in the market soon. As we work with our team and Meta on a daily basis to build the product and customer journey, we think we could be piloting it later this quarter, but just kind of a quick reminder that, given a fairly measured rollout schedule, we won't see asset growth until the back half of -- meaningful asset growth until the back half of 2019, and this product will not meaningfully contribute to earnings until 2020. But we are very excited about where we are with them. Meta is a good partner, and we're excited about where we are with them at this point.

A couple thoughts on credit, from a macro standpoint. Nothing new, but, in and of itself, we think that's good news because the economy in all of our customers and our customers, particularly in the U.S., continues to benefit from the breadth and the strength of the economic expansion. Consumer confidence remains very high, wage growth is good, and larger credit fundamentals have shown no signs of weakening in any material way.

As we called out in our release, net charge-offs on unsecured installment loans have improved, and underlying this, is flat and, in some products, lower first payment default rates, and delinquency collection and payment rates that are higher than forecast embedded in last year. So, in a nutshell, fewer customers are falling behind, and for those that do, more are catching up.

Finally, in terms of legislation and regulation, as most of you have seen, regulations governing small-dollar lending in Ohio will be changing as a result of new law that passed last week and was just signed by the governor. I don't really want to go down the rabbit hole of Ohio politics, but this was a really unusual situation, and the term perfect storm is a pretty apt one here.

As we saw in California earlier this year, where several restrictive measures failed, and we think our industry continues to be very effective at gathering support for a balanced and sensible regulatory environment for small-dollar lending. We're still assessing our options in Ohio, but we currently believe we have until at least May of 2019 to comply. Ohio generated $20.1 million in trailing 12-month revenue for us, but, given the growing book provision build and direct expenses such as advertising, it contributed very little.

Ohio certainly did factor into our growth plans, the old product factored into our growth plans, but we're hopeful that some combination of a reformulated product together with the VERGE CREDIT option from MetaBank for better customers will continue to make us competitive in Ohio and perhaps even increase our market share there over time.

So I'll close where I began, that we're really very -- really, really pleased with the quarter and our progress so far in 2018. We'd like to have put some more money on the bottom line this quarter, but doing so would probably have cost us a lot more down the road. There really is no question that our omnichannel model continues to take market share from legacy branch-based single-pay and installment lenders, and probably even from rent to own and pawn shops.

So we work very hard with our team and with our board to always balance short-term performance targets with longer-term investment and growth objectives. I generally think that we get it right, and I think that the work we're doing developing and transitioning to new products will continue to increase the value proposition to our customers and, ultimately, make us a better company with stronger, more defensible positions in our large and growing markets.

And with that, I will hand it over to Roger.

Thanks, Don, and good morning, everyone. Thanks for joining us. Don already covered Q2 consolidated results and our country-level results at a high level, so I'm not going to repeat, but I will discuss the U.S. results a little more. U.S. revenue growth exceeded 16% on 23.6% loan growth, but year-over-year increase in loss provisions resulted in 5.7% growth in net revenue.

We've discussed at length on prior calls and in our filings the effect of our Q1 -- of our 2017 Q1 loss recognition change on the first half of 2017. Just a reminder, prior to January 1, 2017, installment loans charged off upon payment default -- we changed that at the beginning of 2017 to a more traditional convention. Installment loans began charging off after 90 days of delinquency. So in the first quarter of '17, we had no installment loan charge-offs, since none reached day 91, and in the second quarter of 2017, we had what I'd characterize as undeveloped net charge-off levels because of the transition.

Looking at loss provisions and allowance levels, we've said several times that with hindsight, we were over-reserved for installment at the end of Q1 of '17, and we had to bring most of that over-reserving down during Q2 of 2017. That allowance release had the effect of lowering prior year P&L provisioning versus normalized levels and created a tougher Q2 comp on the provision line.

We also had more sequential loan growth from Q1 to Q2 this year than last year. I'm happy to say that we should expect less noisy comps for loss provisioning in the second half of the year. The growth rate for provision expense should match up better with revenue growth rates for the next couple quarters.

Next, I'll comment on advertising, customer counts and costs per funded loan. Consolidated advertising expense was up 50.8% year-over-year. That's a $5.9 million increase and was 7% of revenue, versus 5.4% of revenue in Q2 of last year. As we think about the rest of the year, we'd expect advertising spend to -- as a percentage of revenue to be slightly higher seasonally in Q2, but in the same range for Q4.

We added 184,027 new customers globally this quarter. That's up 8.6% versus Q2 of last year. Breaking it down, along with related advertising spend, by country, starting with the U.S. -- U.S. advertising rose 57.1% year-over-year. Of this $4.5 million increase, $2.6 million supported the ramp-up of our new Avio installment loans, which we didn't have in Q2 of last year. U.S. new customer counts were up 5.7% year-over-year, fueled by a 21.1% growth in Internet new customers. 50.2% of our U.S. new customers were acquired online this quarter, driven in part by the Avio spend I just mentioned.

U.S. store new customers were down modestly year-over-year, and site to store added 48,000 new customers there this quarter. Because of the Avio and Internet mix shift, U.S. costs per funded was $95. That's up $32 year-over-year, again, because of the mix shift in Avio. U.S. advertising as a percentage of revenue was 6.5%, the range we expected given the ramp-up of Avio and some of the mix shift to online.

Canadian advertising rose 19.4% for three reasons. Number one, mix. We are targeting and acquiring more installment loan and open-end customers than a year ago. Two, the marketing channels we're using. We have expanded cable TV and direct mail spend in Canada. And three, the new product expansion -- supporting new product expansion, including the LendDirect stores in Canada, and increased spend to support the ramp-up of the LendDirect brand.

Canadian new customer counts were down 2.9%, but that comp really isn't apples-to-apples. Last year, a much higher percentage of new customers acquired were 2-week single-pay customers, and because single-pay customers are scored -- because non-single-pay customers are scored more extensively, we do have more declines in the open-end and installment products. As a result, Canadian costs per funded was $87. That's up meaningfully from the same quarter a year ago, but it is down sequentially from $96 in the first quarter and $119 in the fourth quarter of '17.

U.K. advertising rose about $1 million year-over-year, and U.K. new customer counts were up almost 43% year-over-year. U.K. costs per funded was $86. That's up $12 compared to Q2 of last year, but it's flat sequentially pretty much all of this year.

Next, I'll spend a little time covering overall loan growth and portfolio performance. Don already discussed consolidated loan growth. I'll cover a few highlights at the product level. Company-owned unsecured installment loans grew to $179.4 million. That's up $23.3 million, or 15%, versus the same quarter a year ago. Originations increased 7.1% to the same quarter.

Loan growth on the company-owned unsecured installment loans was affected by mix shift from installment to open-end in Canada. We saw a mix shift up there. Unsecured installment balances in Canada were actually down $19.4 million year-over-year. So when you unpack it, the U.S. unsecured installment loans were up almost $36 million, or 35.5%, year-over-year in the U.S.

Open-end loan balances finished the quarter at $91 million, an increase of $64.3 million, or 240%, year-over-year, fueled by growth -- so that's fueled by growth in the season markets, like Kansas and Tennessee in the U.S., where the growth was 26.9% and 20.6%, respectively. We did introduce the product in Virginia, as we mentioned previously, in the third quarter of '17, so we've got that growth, and then in Canada.

Don mentioned that open-end adoption in Canada accelerated this quarter. The open-end balances actually grew $34 million sequentially from first quarter, and it was encouraging that even with this acceleration of the open-end product, Canadian single-pay balances shrank sequentially just $1.4 million.

Moving on to loan loss reserves and credit quality, as I mentioned previously, the 2017 Q1 loss recognition change affected year-over-year comps for installment loan loss provisioning, allowance coverage levels and net charge-off rates. So I'll focus my comments on sequential trends and net charge-off rates, vintage performance and resulting allowance levels. I'll also comment on FPD rates, which obviously weren't affected by the accounting change.

For company-owned unsecured installment, we saw meaningful improvement in vintage loss development and net charge-off rates, which improved 488 basis points sequentially. FPDs were stable year-over-year for this product.

We saw even more improvement sequentially in net charge-off rates for the CSO unsecured installment portfolio, which improved nearly 2,000 basis points sequentially, and FPDs here improved 140 basis points, or 8%, year-over-year.

We are seeing the favorable -- this improvement is attributable to just basically three things. First, origination mix. With our successful credit line increase initiatives, we are loan more like-for-like to our best customers. Secondly, we're seeing the effect of seasonally tighter credit scoring and approval rates last quarter, during tax season. And three, we continue to see general improvement -- continued general improvement in our scoring models. The improvements led to lower required allowance coverage for both unsecured and -- for both unsecured installment products versus the end of last quarter.

For secured installment, net charge-offs were up 282 basis points, while delinquencies improved modestly sequentially. In addition, here, the first-pay defaults improved almost 200 basis points, or almost 12%, versus last year's second quarter.

For the open-end portfolio, first-pay defaults for the U.S. open-end portfolio have improved 50 basis points, or about 2%, year-over-year because of seasoning and improved scoring in our mature states, sequentially improvement in the newer Virginia market, and with the launch of open-end in Ontario, over 56% of our open-end receivables are now Canadian, where open-end first-pay defaults are much better than expected and much lower than in the U.S.

Closing out the discussion of the P&L, our adjustments for adjusted EBITDA and adjusted net income this quarter were pretty much limited to share-based compensation, while the prior year included a litigation settlement. And our effective tax rate for the quarter was 25%, bringing the year-to-date effective tax rate, on an adjusted basis, to 26.5%. We still believe the full year adjusted effective tax rate will be in our previously communicated range of 25% to 27%.

Finally, I'll close with our full year outlook for 2018. Like Don mentioned, we are reiterating -- or we are affirming our full year 2018 adjusted earnings guidance. That's a non-GAAP measure that excludes one-time items, like the aforementioned share-based compensation.

We continue to anticipate revenue in the range of $1.025 billion to $1.080 billion, with continued solid growth in the U.S. and U.K. being offset partially by modest declines in Canada from the additional regulatory changes and mix shift there. In fact, if you do the first half second math -- second half math, at this point, we'd be surprised if we weren't already at the top end of the revenue guidance range.

Adjusted EBITDA in the range of -- our adjusted EBITDA will fall in the range of $245 million to $255 million, and our adjusted net income will be in the range of $110 million to $116 million, with adjusted diluted earnings per share in the range of $2.25 to $2.40.

One other thing to point out about the full year, if you look at our last year numbers, adjusted EBITDA for the first half of the year was higher than the second half of the year. There are a lot of reasons for that, but our guidance for this year implies that that will flip this year. In other words, we reported $124 million of adjusted EBITDA through the first 6 months of 2018. At the midpoint of our guidance, that means we would exceed $126 million of adjusted EBITDA for the second half of the year with related higher year-over-year growth rates.

With that, this concludes our prepared remarks, and we'll now ask the operator to being the Q&A.

We will now begin the question-and-answer session. (Operator Instructions) The first question comes from Vincent Caintic of Stephens.

I was wondering if you could touch on the MetaBank partnership more. At least from the limited press release, it does seem like it's a great opportunity, and particularly when it highlighted a $350 million potential sizing of that. I think that's doubling your portfolio. So maybe -- understanding that the cadence might take a while, but maybe if you could give us a sense of the economics and the potential that could add to your business. Thanks.

I'll just give you a few highlights, and Roger can fill in some of the details. I sort of have to -- my surgeon general's warning, as always, this is -- we haven't launched yet, and I think that, as I said in my remarks, this is likely going to be a product where sort of the rollout and the ramp-up will likely be at a bit more of a measured pace than what we typically have seen. We want to be -- it's a new kind of partnership for us. It's somewhat of a different product. It's, obviously, a line of credit product we'll get a lot of experience with. Given sort of the rate structures built around sort of the amount of times customers draw on the line and how much they have outstanding, the yields are going to be a bit lower. It'll still probably be above what we're getting on our line of credit product in Canada, just as a frame of reference, but lower than what we get in most of the places where we offer a similar product, even an installment product, in the U.S. So we want to be kind of -- we'll be a little bit more measured about letting it build, letting the portfolio build and kind of watching credit as we go along. So having said all that, I think we've made a lot of good progress with Meta. As I mentioned, they're really good partners to work with. We really think they understand our business. I think this is a -- the way this partnership is structured is somewhat unique in the industry, and I think it should work out that both of our -- to both of our benefits. You mentioned the $350 million. So I think we're about $515 million of gross combined loans right now. So if that rolls forward to -- if you were looking at sort of a 2020 thing, it's going to max out on where we are with the $350 million. We'll probably -- if that continues to grow at an above 20% clip, you're going to have -- we're probably going to add another couple hundred million dollars, at least, to that $515 million number. So maybe as we start to add MetaBank, we'll be sort of north of $700 million, and we'll add $350 million to that, so it's probably a 50% build to that $700 million number, so not necessarily a doubling of average earning assets. In terms of the way it's structured, I think we've said low 30s return on assets as this builds. As I said, it will take some time. There's advertising expenses, and as we continue to see the way the provisioning works on this, there will be credit losses that will be kind of in the waterfall that we'll have to kind of work through before it starts returning to either Meta or to us, but, as I said, we think we can get a low 30s return on that earning asset base once it builds up and we get through kind of, I'll call it, the startup and rollout phase of the product, so $350 million of earning assets, north of $100 million of essentially net revenue to us. Again, that's after ad expenses and provisioning. Now, there will be a little bit of associated overhead we have to -- we're doing some of that already, hiring kind of product managers and product specialists to help, and certainly we continue to hire people on the credit underwriting and the IT to help manage this rollout. So there will be a little bit of incremental overhead against that $100 million of net revenue, but, by and large, a very high percentage of that ought to flow to the pre-tax line. And just as a reminder, Meta is going to balance sheet the first $350 million of the program, so there will be no associated interest expense, so it'll be a net revenue share with a little bit of associated overhead, but it should be -- the incremental profitability and the -- it should help our overall kind of profit margin, adjusted EBITDA margins continue to -- and, ultimately, pre-tax and after-tax margins to grow as well. So, I mean, I think there could be a little piece of this that, over time, maybe starts to impact some of our other products, but I think that's -- we've gotten that question a lot. I think it's relatively modest, given where we're targeting this product from a credit standpoint and given that this is a product that allows us to go into a number of states where we don't have options, and in certain states, particularly our two biggest states, Texas and California, it allows us to structure products that are different and fill kind of gaps in the product line up in two very, very big states for us. Roger, anything to add to that?

No, I think that covered -- I think I would just add -- we've talked about it before, but, at this point, we don't see our -- we don't see this product or this launch contributing meaningfully to 2019. It might late in the year. We also don't think -- we certainly don't think it's going to be significantly dilutive to '19 at all, either, because just a reminder that we don't have the loans or the loan losses on our books, and so we share in the waterfall whenever it becomes positive. So we won't see the dilution you would see if we were launching this product on our balance sheet during 2019. And then we -- obviously, I think we've said before we expect it to contribute meaningfully in 2020, and using the numbers that Don just laid out, it's anywhere from $1.50 to $1.70 a share. By then, so if you just kind of roll out the -- again, we haven't even -- I certainly haven't thought about 2020 guidance at this point, and we're not giving 2020 guidance, but if we just look at the Meta program and its potential, I think that's the right range.

I appreciate that. That's helpful to think about it, like $1.50 to $1.70 run rate EPS accretion, so that's really helpful. Thank you. Maybe shifting a little bit more near term -- so I appreciate the updated guidance you gave on 2018 and also the commentary that maybe there's a good chance that you might come out ahead of that. If you can maybe -- kind of what assumptions on maybe the low end to the high end of the EPS guidance range, and what would it take to get maybe beyond that, and then also if you could help just -- since you're relatively new, if you could help us understand the cadence of what happens in the third quarter and the fourth quarter in terms of seasonality. Thanks.

This is -- I mean, I'll kind of take the first half of that and let Roger fill in some details on the -- the first half of your question and let Roger fill in some details on the second half of the question. Just if you look at the midpoint of our guidance range, if you look at the top line, it's 1052 in total revenue. The first half, as I mentioned, we did $510.7 million, so we really -- and I think that -- and the first half of the year, we grew 15.7% top line year-over-year. Our expectation is that our revenue in the second half of the year will -- and I think we grew 14.8% for the quarter. It'll probably come in somewhere in that range for the back half of the year. And if you just do the math out on that, that gets you to sort of -- it gets you to about $1.1 billion, so it's even above. As we mentioned, it kind of gets you above the top end of the guidance range. Now, again, there's -- as we mentioned, there's the pretty meaningful -- 2/3 of Ontario is a big piece of -- Canada last year, just in context -- and, again, we give you the full segment details there, but Canada last year did -- this is in U.S. dollars, $186 million of revenue. We think that's still going to grow. The top line in Canada is going to grow about 7%. When you're talking about transitioning 2/3 of that P&L from a product standpoint, I think that just gives you a sense of why we're just a little -- we just want to make sure we're just a little cautious in terms of -- we're really happy with where things are going, but we don't want to spike the ball at the 10-yard line here in terms of overpromising and overprojecting on what's happening in Canada, other than to say we're really happy with it and our team has done a great job. So that's kind of the top line. We would also expect -- Roger, we would expect that our provision in the second half of the year would run about with revenue.

It'll be close. It will be much closer, much less noisy than the first half.

Right. Exactly. So, again, in the context of the first half of the year, revenue grew 15.7%, provision grew 36%, so if revenue is going to grow in that 15% range, 14%, 15% range, provision grows in that, so net revenue growth should run about with gross revenue growth. The other, bigger piece of the P&L, from an advertising standpoint, given the continued transition in Ontario -- there's some ad support for that. Again, that's transitioning existing customers, but we are getting a really good take-up rate from new customers, so we're going to continue to advertise. As we like to say, if you live in Ontario, it's very -- unless you're kind of living in a cave, we think we've reached you multiple times with this advertising, but we're going to continue to push that. So we would expect -- in that and our Avio product in the U.S., we would expect that our ad spend as a percentage of revenue -- last year, I think it was about 6.5% in the second half of the year. It'll likely go up maybe 140, 150 basis points, in the 7.8% to 8% range, for the back half of the year. So that'll be a bit higher than last year, but, again, the revenue growth is there and the provisioning should be in line with revenue. And the rest of it, we'll see some G&A growth from public company costs, et cetera, but the rest of it we always talk about is there are call centers and stores, and we just aren't seeing a lot of increases in the growth rate or the expense line there. That's just some high-level thoughts about why we think we're pretty optimistic about the back half of the year, and Roger kind of addressed the way -- kind of the earnings progression.

Yes, good point on the newness and the seasonality. As you guys know, the first quarter is very high seasonally in terms of earnings and adjusted EBITDA. The second quarter is the low point of the year seasonally, and then we kind of build back. Q4 gets up a little -- Q4 seasonally should be in the same range or higher than Q1 seasonally, and then the third quarter is kind of right in the middle. So if you look at our guidance and think about that seasonality, you get -- the fourth quarter and the first quarter are the two highest quarters of the year, Q2 is the lowest, and Q3 is just kind of right in the middle.

The next question comes from Bob Napoli of William Blair.

I guess the Canadian -- the conversion of the Canadian business, Don and Roger, I mean, that is pretty -- what gives you the confidence this early on? You're ramping it up at a much faster pace. What gives you the confidence that that business is going to perform as you expect from a credit perspective? And why has the single-pay held in, as well? Do you expect that to drop off more aggressively?

Bob, this is Donald. So we ran a pilot in Windsor, Ontario, that we started -- as I mentioned, we started in February, and so it was sort of big enough to be kind of meaningful as a sample size and a learning exercise, and we saw a really good take-up rate, good conversion rates from existing customers, and, as I mentioned, a really healthy number of new customers. And if you look at our revenue in Windsor in the second quarter, those stores, it actually grew -- the total revenue base grew there versus the prior year. The first-pay default rates on the new product are in the 6%, 7% range, which is really, really good. Again, you've got a product that's going to be a meaningfully lower APR rate, so, as we like to say, credit has to get better. It's not just an academic exercise. I mean, the credit has to get better or the margins are going to get squeezed. So the product performance has been -- the credit performance has been really good. The acquisition costs -- again, you've got an installed base of stores, you've got good local awareness in the brand, and we've been advertising since we bought that business in May of 2011. We've been advertising really consistently there on TV. But we also continue to get -- I think we had over 15% of our new customers in Canada in the quarter that came online, which is a -- that percentage continues to tick up. So the omnichannel side of the business and the advertising benefits you get from the omnichannel approach I think is really starting to take hold. In the U.S., we're getting -- 2/3 of our new customers are coming online in some way, shape or form. We've got a long way to go to catch up with the U.S., but I think that it's not just a -- the numbers look good, but the operational side of it we feel really, really good about, and I think that's a testament to the work the people in our branch and our call centers have kind of always done with customers, and I think we've -- if we have the time to communicate new products and services and new options to customers, I think exhibit a pretty high degree of trust in what we do for them, and I think they understand and appreciate the value proposition in the line of credit. It is more credit at a cheaper rate and more flexible repayment terms, and it fits kind of well in their budget, and they use the online tools and the mobile tools to help them kind of manage that, so I think it kind of touches all those bases.

Yes, and, Bob, I would add that we started doing installment loans in Canada at the beginning of 2016, and so we -- obviously, we have that experience, and if we match that up with our open-end experience in the U.S., where we've been doing open-end for over 5 years -- if you match all that up, we expect the open-end to perform a little bit better, perform better than the installment, and so far that's exactly what we're seeing.

And just one more comment, just for the context. As I mentioned, we did $186 million in revenue in Canada last year, U.S. dollars, and our forecast is we'll do right -- for the full year this year, we'll do right around $200 million, so we'll see some revenue growth, but adjusted EBITDA last year in that business, in U.S. dollars, was $54.6 million. We could see in the range of a $20 million reduction in that adjusted EBITDA number. We initially thought it was going to be down in the $10 million range. It could be more than double that down, given the acceleration of the transition. So that's a really meaningful hit on the earnings side, but the flip side of that, as I mentioned, if we get through this transition, the exit rate should put us on a path to get back to that 2017 EBITDA number. Now, this is not a forecast, I just want to make it very clear, but that's -- when we talk about objectives internally and think about things, it's our idea that we're going to get back to that 2017 adjusted EBITDA number. We may not get all the way back there in '19, but we're certainly going to be -- we'll get a large measure of it back. And when you look at that from an earnings growth standpoint, and fortunately, the U.S. side of the business is doing well enough for us to be able to sort of, I guess, afford to do that transition and still have our overall numbers be really, really positive.

The other part of my question was why did the single-pay, and then I'll turn it over -- why did the single-pay hold up as much as it has, and do you expect that to -- the decline of single-pay to accelerate?

Hey, Bob. It's Bill Baker. I mean, when we ran the test, we did 21 testers in Ontario starting in February, and what we experienced there with single-pay is we did see a decline due to all the things that Don just talked about with the line of credit, but we did see it start to come back, and we expect to see the same thing in the broader Ontario rollout. We actually built the model to allow for that, and so what we want, the idea is to provide the line of credit, which is a much larger loan, lower yield, and then still allow for the payday product for kind of the emergency use for a segment of customers that were very carefully scored and kind of registered in the system. So we expect that to come back a bit, and I think that will temper the decline on single-pay, but clearly there's a big transition, but we -- if you look at the yields, even with the extended payment plan and the net income limitations, it still yields around 200%, so it's an extraordinary viable product, and we don't necessarily want to be out of that business in Ontario, so it's a bit of a balancing act to make sure we offer the right product to the right customer.

The next question comes from Moshe Orenbuch of Credit Suisse.

Most of my questions have actually been asked and answered, but could you talk a little bit about the yield on the open-end product and how that's going to develop over the next couple of quarters?

For the Canadian product, that product is regulated nationally in Canada, and the way it all works out against those national regs is the interest yield is about somewhere around 47% to 48% annually, and then we offer optional credit protection insurance, and about 70% of the customers buy that. It's a real insurance product. It's a good product. About 70% range, 3/4 of the customers, buy the credit protection insurance, and that adds about 12 points of yield. I'll put yield in quotations marks. So you wind up -- the portfolio winds up adding, on a revenue basis, about, I don't know, low -- right around 60% to 62% on an annualized basis. And then the U.S., obviously -- just to reiterate, the U.S. open-end products are regulated at the state level, and the rule of -- the regulations there generally are something in the range of 0.8% per day of interest, so a much higher-yielding product, but with much different credit performance in the U.S. than in Canada.

The next question comes from John Hecht of Jefferies.

A lot of my questions have been asked. First of all, at the product level, it looks like, domestically, the unsecured product has a lot more demand and good credit. I'm wondering if you can talk about what you're seeing in terms of the customer level demand patterns, and is there any geographic area where you saw more of that than you expected?

Hey, John. It's Bill Baker. So I think the demand on the unsecured product is really almost more of a testament to our underwriting and scoring, just to the fact that we can offer customers more money with more confidence than we did 5 years ago, for example, so they don't need to bring a title to do a secure loan. We just have better scoring, better underwriting, better credit data, and so we can offer them more money with an unsecured product, and I think that's really a big part of what we see in California and Arizona. You still see demand for it, but it's growing slower because I think of what we've done. I don't think it's necessarily a market issue. I think it's more of a -- we've just gotten more confident. I do think you see more confidence in the customer in places like California and Arizona, where -- and Nevada, for example, where you see incomes higher, minimum wages going up at a fairly drastic rate, and people just have more secured assets to pledge, quite frankly. I mean, we're seeing growth in Arizona that we haven't seen in years, for example, which is basically a title state. So it's a bit of a mix of I think what we've done and also what's happening in the market.

And then, second question, thinking about the omnichannel, any change in composition of how consumers are finding you guys and reacting to marketing, starting online or in the store, and how you close them out as well?

I think as we disclosed in the release, it's about -- it just about 51% of people are finding us online. Again, some of those go to the store, but most of them close online. I think it's fairly consistent, but we continue to see it leaning towards more and more online, although we're extremely happy that we have 214 branches in the U.S. that we can funnel people to. I think Roger mentioned it was -- he mentioned the number in his statement. And most of those customers we could not have closed online, so the site-to-store program remains extremely viable. I won't get into a ton of detail, but we have a number of channels that we're testing right now that are really positive, and I would say the thing about that is those channels are -- from our analysis, they're unlocking a new customer. So we see very little overlap with the internal database that we have of over 70 million applications, so we are -- I think we're very bullish on the back half of the year with new channels and a really efficient cost of funded.

This is Don. I'll just add to that, that if you look at our store base, not only do we think -- it performs very well. They're very large-format stores, really high-margin stores, and we've really -- Bill and his team have really worked to sort of reorient those stores to be part of an omnichannel program where customers we couldn't close online can feel comfortable bringing their information into the store and meeting somebody at the store and closing a loan there. And if you think about that store from sort of like a positioning and a share standpoint, if you look at just in the last 5 years in California, the number of branches has fallen about 19%, and in Texas, it's even more pronounced. It's fallen about 32% in the last 5 years. And, actually, that's kind of accelerating, so you're talking about 1,000 stores have closed in Texas, so from 3,000 to 2,000, and in California, it's gone from somewhere like 2,200 to 1,700. So our stores not only are good in and of themselves, they're sort of standing out and more viable from a share standpoint and an awareness standpoint given what's gone on in the overall industry.

The next question comes from John Rowan of Janney.

Don, just to be clear, the $100 million figure that you gave for the fall run rate on the Meta deal, that's pre-tax, correct?

That'll be pre-tax, John, yes.

And then the bonds are still well above par, but they are below kind of the peak that we've seen over the past few months. Does that change your appetite to possibly refinance those?

I think what they're doing, John, I think is they're kind of trading to the -- if you look at the first call, which is in March, we had -- and the call premium is 1 plus half the coupon, so it's a 106 call in March. So I think you kind of look at what you get between now and a 106 call, I think they're kind of trading towards that. Again, I haven't looked at them in the last 10 days or so, but I think they're trading closer to that line, so I think we still have a big appetite to do it. It just becomes a question of when to do it. Do you want to do it sooner and, obviously, pay some make whole, or do you want to wait for the call date, and I think that just becomes an exercise on where do you think you can do -- where do you think the new bonds are going to price out? So our idea here was to kind of get through the quarter, get through, obviously, reporting, et cetera, file the Q and all that kind of stuff, and then we'll have some more bandwidth here to take a harder look at the fixed income market. But we'll be doing a lot of that work between -- really, over the next 6 months. I think we still have a very strong appetite. It's just a question of where we can get them done right now.

And, Don, while I have you going, can you remind me, is there a legislative session for California in the fall? Just remind me of how the seasons run.

California still has -- and I'm not an exact expert at this, John, so I'm going to -- you may just follow up with our GR folks, but they do have some limited session in the fall. But, really, given the way that -- if bills don't cross over from the assembly to the Senate basically by Memorial Day, then they have to start over again next January, and so the whole process starts again in January.

This concludes our question-and-answer session. I would like to turn the conference back over to Mr. Don Gayhardt for closing remarks.

Great. Thanks, everybody, as always, for taking the time to join us today. We will look forward to talking to you again after we report our third quarter.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.