Credit Acceptance Corp (CACC) 2016 Q4 法說會逐字稿

Good day, everyone, and welcome to the Credit Acceptance Corporation's fourth-quarter 2016 earnings conference call. Today's call is being recorded. A webcast and transcript of today's earnings call will be made available on the Credit Acceptance website. At this time, I would like to turn the call over to Credit Acceptance Senior Vice President and Treasurer, Doug Busk.

Thank you, Brian. Good afternoon, and welcome to the Credit Acceptance Corporation's fourth-quarter 2016 earnings call. As you read our news release posted on the investor relations section of our website at CreditAcceptance.com, and as you listen to this conference call, please recognize that both contain forward-looking statements within the meanings of federal securities law.

These forward-looking statements are subject to a number of risks and uncertainties, many of which are beyond our control and which could cause actual results to differ materially from such statements. These risks and uncertainties include those spelled out in the cautionary statement regarding forward-looking information included in the news release. Consider all forward-looking statements in light of those and other risks and uncertainties.

Additionally, I should mention that, to comply with the SEC's Regulation G, please refer to the financial results section of our news release, which provides tables showing how non-GAAP measures reconcile to GAAP measures. At this time, Brett Roberts, our Chief Executive Officer; Ken Booth, our Chief Financial Officer; and I will take your questions.

Thank you.

(Operator Instructions)

David Scharf, JMP Securities.

Hi, good afternoon, thanks for taking my questions. Wondering if we could start just addressing volumes and demand to close out the year, as we think about the dealership count holding steady from the third quarter? Setting aside the usual seasonal drop off in Q4, it looks like average volume per dealer declined a little more.

Just wondering, as we think about some of the volume metrics, are the biggest factors affecting it competitive terms from other lenders? Or are they independent dealers perhaps losing shared franchise dealers? Or is it, thirdly, more consumer-related? Any sense that we are starting to see a peaking of consumer demand?

I think the first factor I would look at is the competitive environment. That certainly makes things challenging. We're not getting any improvement there, but I do not think it got any worse either. So I think what you saw in the fourth quarter is -- as we talked about in prior calls, our strategy when the environment is competitive is to focus on growing a number of active dealers. Difficult to grow volume per dealer when it is very competitive.

But as I think we alluded to several quarters ago, as the base of dealers gets bigger, it becomes more difficult to grow that at a fast enough rate in order to offset the decline in volume per dealer, and to a lesser extent, attrition. So that is what we're seeing right now. It is just, we're having difficulty signing up enough dealers to offset those other two factors.

I think in the fourth quarter, the other thing that came into play is, we addressed some of the negative variances that we have seen in recent originations. When we do that, I think it causes our initial collection forecast to decline, which means we advance the dealers less money. So that has a one-time impact on volume per dealer, and also in attrition.

Got it. And Brett, your comment about, is this the size of your dealer count -- it gets larger, it gets harder to grow off that base. Should we think be thinking about, call it the 7,500 active dealer figure? Are we bumping up against a ceiling, or is that just a reflection of the competitive environment? And if competition were to ease and dealers had fewer funding options, is there still a lot of upside to that 7,500 number?

I think it is too early to call it a ceiling. It's definitely a point of resistance. If the competitive environment, obviously if that changes, that will change everything. But if we assume that the current state continues for the foreseeable future, I would look at it at this point as a point of resistance, and not a ceiling.

Got it. Just a couple more. Just curious, switching to the provision side, it looked like the allowance picked up a bit to close out the year. I think it was 7.6%, and you already noted some of the collection patterns. As we think about modeling 2017, I know you do not give explicit guidance, but is that high 7% allowance rate a reasonable target for us?

As I think you probably know from prior calls, we tend to look at the adjusted numbers internally. The best number in the release, if you want to sort of get your arms around the economic impact of forecast changes, on the second page of the release, the net cash flow number declined $14 million for the quarter. That is on a base of $5 billion or so in un-discounted cash flows. So a very modest decline in the quarter, $14 million, that is really the economic impact.

And then as you know, the way that GAAP handles that from provision standpoint, I think it makes it difficult to see the economics. The adjusted numbers we provide, we can do the best job of sorting all that out and giving you a number that you can focus on.

Got it. And one last one, and I will get back in queue. It looked like your spread actually picked up a bit sequentially. Obviously you're pricing your advanced levels in accordance with maybe a revised thinking about collection patterns. Is the 21% to 22% spread generally based on your internal models and your target returns? Should we be thinking about that as, in effect, a floor, at which you wouldn't underwrite if the spread was below that? I mean, we don't have access to obviously all your loan-by-loan return targets, but is this a good way of thinking about a floor, in terms of the spread over the advance rate?

No, I don't really have any guidance on that number. As we've talked about in prior calls, the way we price is, we try to maximize the full economic profit we generate. So unit volume times economic profit per unit. Then we make adjustments all the time based on the model that we run. At times, that means we get a little bit more aggressive and advance a little bit more; at times, it means we do the opposite. But that is how we price. We don't price based on that spread.

Right, got it. Okay, thank you.

You are correct. It was a favorable change in Q4 with respect to --

Yes, I know. We're obviously looking for shortcuts in terms of how to model it, using it as proxy for yield. But, understood. Thank you very much.

Yes.

John Rowan, Janney.

Good afternoon, guys. Going back to that commentary on volume, so unit volume was down 6% year over year. Dollar volume was up, because the average loan has bumped up near $20,000, which is considerably higher over the last couple of years. You know, if we're getting to a point where there is resistance in this current environment at the current dealer partner number, at what point do we start to see the loan portfolio actually contracting? Because I would think unless we continue to see very big increases in duration and average loan size per consumer, with unit volume coming down, we're going to have to start seeing some contraction in the loan portfolio at some point in the near future.

It just depends on your assumptions for unit volume going forward. Obviously the loan portfolio is growing pretty rapidly right now, and the dollar volume increased in Q4. So I think you're a ways away from contraction. But it just depends on what assumptions you make regarding unit volume going forward.

But is that increase in loan growth mostly a function of the loan portfolio being -- I mean, you have it in the press release here. The almost $19,000 average loan per consumer, which is up from $16,000 in 2015. Is that really what is driving a lot of the dollar volume gains on a per-dealer basis, as opposed to the increase in dealer partner numbers?

Yes, I think you're right. Unit volume declined 5.6% for the quarter. So to the extent there was a increase in dollar volume, it was all related to a change in the average size of the transaction.

Okay. What was the increased legal fees that you guys noted in the press release?

Nothing specific there. You know, we just have a bunch of issues ongoing that are consuming increased amounts of legal resources.

Okay. And then the 11.7% return on capital that you noted. Just historically speaking -- I only have a few years here -- but is that the lowest you guys have seen in a while? Maybe if there was another point in time where that number had been that low, kind of compare where we were in a competitive cycle versus then?

From memory, our returns were in the 11% range back in 2007 or 2008, I believe.

Okay.

On an annual basis, we were low 11%s, and 2008 was the lowest it has been. Obviously 2001, 2002 and 2003 were single digits. But since 2004, this was -- we were in the 11% range in 2008, and then again this year.

Okay. And then just go back to the dealer partners, you guys give some numbers on the new dealer partner participation -- you know, consumer loan volumes from the new active dealers. And those were all pretty solidly negative. Again, is that just a point of this resistance level that you're talking about in the dealer partner program in the current environment, just given that you're just not offsetting any attrition, and that all those numbers are down pretty solidly?

I think the volume per dealer, I would attribute that mostly to the changes we made to address the negative variances we are seeing in the lease origination.

So the decrease in the new dealer partner volume -- because consumer loan volumes from new dealer partners were down 30% year over year. Is that because of pricing changes in the fourth quarter, or is that because we're hitting a resistance level in the dealer partner base?

Well, I think, two things, if you just focus on new dealers. Because the changes we made that I just talked about related to the variance, and we signed up fewer new active dealers this quarter than we did the same period of last year.

But there have been no changes in the salesforce?

No.

Okay. And then just one other thing. I noticed the last few years, we have seen this shifting of volume out of the portfolio program and into the purchase program. And it looked like this quarter, we're actually starting to see that turn around a little bit. Maybe give us any indication why we started to see more volume go back into the portfolio program?

I mean, the change was pretty modest. It was 23.8% in unit terms last quarter, and it was 21.4% in the fourth quarter. So a couple of percentage points is not a really big change.

Okay. Thanks for answering my questions.

Moshe Orenbuch, Credit Suisse.

Great, thanks. Can you talk a little bit about the changes that you made in the forecasting process? Like what new variables or data do you have? It is interesting, because the earlier periods, I mean, two of these three years, you had substantial portions that had already been paid back. So I guess, how does that -- I mean, does that have a noticeable impact? Could you just talk a little bit about those changes?

Yes, we periodically refresh the model, so it is based on all the information we collect, both at loan origination as the loan moves through the servicing process. So it is really just mostly an update of the data, a refresh there. And also we redo the actual scorecard and re-weight the variables based on seems to be most predictive. So it is something we do periodically. It is more, I think, a routine update. And really, the numbers were not all that different from what we had before.

Got it. And you had mentioned that your newer loans have a lower advance rate. I guess, can you talk a little bit about how that has had -- I should say, if that has had an impact as you go to market, to dealers? How should we think about that? And is that something that is likely to continue, does it level off, does it turn around?

That is very difficult to say. I mean, we react to what we have seen so far. We've got some negative variances that you see in the table that we provide, and we obviously look at those as well, and we have to react to those. We do not want to see negative variances continue. So as we make changes to adjust for that, that has a negative impact on loan volume.

Okay, all right. And then just, is there an impact on from the perspective of the way you think about the cash flows, with respect to dealer hold-back when a dealer attrites, if a dealer leaves and has kind of earned some of that?

Our contractual arrangement with the dealers is, if they become inactive prior to doing 100 loans with us, then they forfeit their rights to that dealer hold-back. Other than that, it works exactly the same. We continue to collect and pay back the advance, and remit the whole back of the dealer in the normal course. So whether they are active or not, we still continue to service the loans and pay them the hold-back that they are entitled to.

Right. And I guess I'm just trying to get my head around if there are any other impacts from dealer attrition rising. That was one of the things that kind of jumped out that could happen. Are there others that we should be aware of?

No, other than the obvious. I think the obvious impact is loan volume. Which, we prefer to see attrition lower. That's what we are focused on. We would like to see the dealers continue with the program for as long as possible. But I think the primary impact is just on loan volume.

And at what point -- it looked like you bought back a modest amount of stock in the quarter. I mean, is there any kind of thought as to how we should be thinking about the interaction between the capital you are generating and how you deploy it?

I mean, we bought back about 450,000 shares of stock during the quarter, total cost of about $80 million. I do not think we really have any change in the way we are thinking about that. We've talked about how we approach that decision in the past, and our thinking has not really changed there.

Got it. Thank you.

John Hecht, Jefferies.

Actually, I think all my prepared questions were answered. But a couple that just came to mind is that, you guys gave us the -- if I recall, you give us the October volume, or how it was trending when you last reported. Can you tell us like November, December and January, just to give a sense for more recent volume trajectories?

Sure. In unit terms, just as you mentioned, October was down 8%; November, down 5%; December, down 2%. January is not complete, so we are not going to count on that. But that gives you the trajectory during the quarter.

Yes, thanks very much. And then maybe because there has been a lot of increasing investor focus on residual values and used car price trends and so forth. I mean, you give any buy substantially different than others do, and you are buying at a big discount rate. Can you give us a sense of how to think about sensitivity to declining residual values? Or is your sensitivity more -- I guess, is it more to frequency or severity? And if you are to the extent you are exposed or sensitive to declining used car prices, can you quantify that? Is it materially meaningful, or is it something you really count for when you're buying a loan?

Yes, I think we talked about this last quarter. The way we look at it is, when we finance a used vehicle, that is a depreciating asset, so we just focus on and track the depreciation and book values over time. And we have a metric to look at it every month on that. As many people have written about, vehicles are depreciating faster this year than they have, I think faster than any time back to 2008. I think 2008 was a bit worse than 2016, but not much.

So in terms of vehicle depreciation, we are at the unfavorable end of the spectrum compared to where we have been in the last 12 years. The fourth quarter was worse than the full year. Vehicles depreciated faster in Q4 than they did in Q1, Q2 and Q3. But the magnitude isn't something that we really worry about that much.

The difference between -- I think there was one year out of the last 12 where vehicles hardly depreciated at all. I think it was 2011, if you look at it on a trailing 24-month basis. So what that means is, the contracts we wrote in 2009, those vehicles, by the time we got around to selling them, to the extent we repossessed them, they were worth about as much as they were when we go for a loan -- which is unusual.

Today you are in a much different position to that, where over 24 months, the vehicle has depreciated quite a bit. The difference in terms of the highs and lows is about 300 basis points in terms of the collection rate. So it's significant, but that is between -- over the last 12 years in the absolute best market we have been in and the absolute worst market we have been in, it's about a 300-basis point change.

Now, unfortunately, we cannot really predict what it's going to be over the next 24 months. So even if we wanted to, it would be difficult to build that into our model. As we've talked for the past, the way we address that is, we shoot for a return. That means if the loans underperform our forecast, it still means the business we wrote was very flat. We would prefer to address it that way than trying to be experts in predicting used car values over the next 24 months.

Great, I really appreciate that color. Thanks.

Robert Dodd, Raymond James.

Hi, guys, this is actually Leslie Vandegrift this afternoon. Quick question on the forecast change for fourth-quarter 2016. In the press release, you guys split it out for the first three quarters, and then the fourth quarter. And the initial forecast for fourth-quarter vintage was 53.7%, and it increased up to 65% by the end of the quarter. Can you talk to me about how there is that significant change in that three-month period?

I would rather have a positive no there than a negative number, but early in the loan life, I think there was some variation there. So I wouldn't get too excited about that number until we get another quarter or two under our belt.

Okay. So it's more of what is coming in, rather than actual changes to what you were first thinking about it?

Well, I think just early in the loan's life, when we have an initial forecast, when we pay the dealers the advance, and then one month in, we adjust their forecast based on whatever happened in the first month. Mostly, did they make their payment or not, and any change in the value of the vehicle. But there is quite a bit of volatility -- there is more volatility in a collection forecast at the start, and then it starts to level off and be a little bit more stable. So given those loans are only at the most, three months old, and at the least, one month old, I just wouldn't put a lot of emphasis on that number.

Okay, all right, thank you. And then on Chairman Foss's retirement, I know in the original press release, there wasn't an announcement of someone replacing him. Do you guys have -- is there a plan for that to occur, for that position to be refilled?

No, there's no plans at this point to have a Chairman. We have a Lead Director. But we do not need a Chairman, and we have no plans to replace him.

And then, do you have an estimate for the amount of shares that he was beneficiary of, that he sold back in 2016? I know that there was the $500,000 that were registered, but we cannot see when those sales occurred.

Don would file Form 4s when he sold shares. So those Form 4s would be out there. From memory, I don't believe he sold off any material shares of stock during 2016.

Okay, all right. That is all for me. Thank you.

(Operator Instructions)

Clifford Sosin, CAS Investor Partners.

Hi, guys, thank you. Can you characterize attrition among your more seasoned dealers -- dealers who have done at least 100 loans or dealers who have maybe been around for some period of time, however you want to define the cut-off? Not just the level of attrition, but the trend -- has there been a meaningful step-up any attrition amongst your more tenured dealers?

I do not have anything to add there. Attrition is up. I am assuming it is probably up for both. But there is nothing that has come to my attention that would give you any more color on that.

Okay, thank you.

Brian Gustavson, 1060 Capital.

Hi, there. I'm just looking at the initial loan term. It has gone from, call it, 47 to 53 in the last three or four years. And I am wondering what your thoughts on that? Does that keep on ticking up or do you see that coming down? And do you think that might be a reason that collections have ticked down more recently? Thank you.

We write terms all across the spectrum, from six months to 72 months, so the average term is the function of the mix. We do not have any current plans to increase the maximum term beyond 72. So the average will just be a function of our pricing strategies and where we see the most opportunity. I didn't have a forecast there.

But you know, as we've talked about in prior calls, we, over a period of time, when I first started, we would only write a 24-month loan. And we accumulated data and extended it out to 30, and accumulated more data, and we walked our way out to 72 months over a period of about 25 years. So we have been careful about it. And we're comfortable with the terms that we offer now.

For a given loan, if you do not change anything else and you increase the term, the collection rate is going to be lower. But we build that into our forecast. So that is not necessarily the reason why you saw the variances that you did in the 2015 and 2016 originations. I do not think it was related to term -- more related to just the overall market, adverse selection, the decline in used car value that we talked about.

Got you. Thank you.

Thank you. With no further questions in the queue, I would like to turn the conference back over to Mr. Busk for any additional closing or comment remarks.

We would like to thank everyone for their support and for joining us on our conference call today. If you have any additional follow-up questions, please direct them to our investor relations mailbox at IR@CreditAcceptance.com. We look forward to talking to you again next quarter. Thank you.

Once again, this does conclude today's conference. We thank you for your participation. Have a good day.