Good morning. My name is Robin, and I will be your conference operator today. At this time, I would like to welcome everyone to the Q3 FY 2014 Conference Call. [Operator Instructions] Ms. Katharine Kenny, you may begin your conference.

Good morning. Happy holidays. Thank you for joining our fiscal 2014 third quarter earnings conference call. I have with me today Tom Folliard, our President and Chief Executive Officer; and Tom Reedy, our Executive Vice President and CFO. Before we begin, I need to remind you that our statements today regarding the company's future business plans, prospects and financial performance are forward-looking statements that we make pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995. These statements are based on management's current knowledge and assumptions about future events that involve risks and uncertainties that could cause actual results to differ materially from our expectations. In providing projections and other forward-looking statements, the company disclaims any intent or obligation to update them. For additional information on important factors that could affect these expectations, please see the company's annual report on Form 10-K for the fiscal year ended February 28, 2013, which is filed with the SEC. Before I turn the call over to Tom, I would like to remind all of you that our next regular Analyst Day takes place on January 21. If you're interested in attending, please let us know. Tom?

Thank you, Katharine. Good morning, everyone. Thanks for joining us today. Used unit comps for the third quarter were up 10% compared to last year despite the toughest comparison of the year, primarily due to better conversion, as well as a modest increase in traffic. Total used unit sales grew by 15% in the quarter. Used vehicle gross profit grew by 16%, and used vehicle gross profit per unit of $2,149 was generally flat compared to last year. Wholesale unit sales, up 4% due to the growth in our store base. Our wholesale gross profit was similar to the prior year, as the 4% growth in units was offset by a decrease of $36 in gross profit per unit. Extended service plan revenues were also similar to the prior year, as the reserve adjustment related to the increases in cancellations offset the effect of our sales growth. CAF quarterly income increased 16% to $84 million. The bottom line is we're pleased with another strong quarter. Net earnings grew 12% to $106.5 million, and net earnings per diluted share rose 15% to $0.47 per share. I'll now turn the call over to Tom Reedy to talk about financing. Tom?

Thanks, Tom. Good morning, everybody. In the third quarter, CAF income grew 16% compared to the third quarter of FY '13. Average managed receivables increased 24% to $6.8 billion, and portfolio growth continues to be largely driven by strong origination volume. Similar to the first half of the year, managed receivables grew at a faster pace than CAF earnings, as the increase in loan volume was offset by compression in the spread versus last year's third quarter. Weighted average contract rate for accounts originated during the quarter was 7% compared to 7.7% in last year's third quarter, but up slightly from the 6.8% we saw in Q2 of this year. The allowance for loan losses increased to $68 million, and at 1% was flat as a percentage of managed receivables. Credit losses in the quarter were moderately better than our expectations. For CAF, net loans originated in the quarter rose 12% to $961 million, and net penetration was 41% for both this and the prior year's third quarter. Third-party subprime providers accounted for about 18% of our sales in the third quarter compared to 15% in last year's third quarter. As you know, we have experienced an increase in subprime volume over the last couple years, as third-party providers have made more attractive offers to customers. Late in the quarter, we began to see them tighten the credit offers that they've been providing. Finally, we plan to launch a test during the quarter to actively learn more about originating and servicing customers who would typically be financed by our subprime providers, an initiative we've been working on for over a year. Our objective, at this point, is to gain familiarity with the customer segment and to determine if it is appropriate for CAF to participate as one of the lenders in this space. Beginning in the fourth quarter, we plan to route a small percentage of these applications to CAF, and we'd expect to originate approximately $70 million over the course of the next year. This represents less than 2% of what CAF originated from -- in the last 12 months, and we plan to fund this test separately from our securitization vehicles. Tom?

Thank you. Regarding our sales mix, there were very few changes of note. Similar to the second quarter, sales of compacts and midsized vehicles grew a few percentage points. As we've discussed before, our mix of vehicles will vary based on customer demand. Total SG&A for the quarter increased by 11%, reflecting the 12% growth in our store base since the beginning of last year's third quarter and variable expenses related to higher sales. SG&A per retail unit was down $98, again, largely driven by the 10% comps. In the third quarter, our average monthly web visits neared 12 million hits per month, up 36% compared to last year's third quarter. This quarter was the first during which more than 50% of our traffic came from devices other than desktops or laptops. Average monthly visits to our mobile site represented about 26% of total visits, and visits utilizing our iPhone or Android app represented nearly 12% of our traffic. Also during the third quarter, we opened 3 stores: 1 in Jackson, Tennessee; 1 in Brandywine, Maryland; and 1 in St. Louis. And after the third quarter ended, we also opened our second store in St. Louis and 2 stores in our Philadelphia market, including our King of Prussia store, which will open today. We announced 4 planned openings for next year's third quarter. Three are in new markets for CarMax, including Portland, Oregon, Tupelo, Mississippi and Reno, Nevada. We will also open our third store in the Raleigh market. And with that, we'll be happy to take your questions.

[Operator Instructions] And the first question is from Simeon Gutman. Simeon Gutman - Crédit Suisse AG, Research Division: So I have one and a follow-up. The first might have a couple of parts to it. On the finance side regarding the subprime tightening, can you talk to -- is the number of offers being made is going down or is it the same number with higher rates? And can you discuss what's sparking the change and if you're seeing any other changes in the actions of other lenders?

I'll take that, Simeon. As far as what tightening means, it really means they're going backward on the changes that they've made over the past couple of years, which is downpayment and kind of ease of documentation. So it seems to slide back a little bit on that. As far as where it lands, we don't know, it's too early to tell and what that means. And as far as why they're doing it, they run their own portfolios, just like CAF does, they're managing to optimize profitability and size as well. I can't speak for them, but I guess -- I would guess that they're taking moves to manage risks based on what they've been seeing over the past couple of years. I do not believe that it's a CarMax-specific issue, though. Simeon Gutman - Crédit Suisse AG, Research Division: And then the other lenders at the other part of, I guess, the portfolio, not the non-CAF stuff, some of the other lenders, are you seeing any changes there?

Actually, this quarter has been a little bit positive vis–à–vis where they've been historically. Simeon Gutman - Crédit Suisse AG, Research Division: Okay. And then the one follow-up is in, in testing your own portfolio, is that something that was discussed with your current partners and might you expect a change in the fee you pay as a result of that test?

No, this is something that we've made them aware of. In fact, we've been working with them over the past year trying to do due diligence and get smarter about the -- about working in this space, so it's not a surprise for them. And the nice thing about being back in growth mode again is that we can grow the portfolio and allow partners to continue to grow and bring new people in, like ourselves, if that's what makes sense.

And your next question is from Matt Nemer.

I'm just -- just a follow-up on subprime. I'm curious if your in-house auctions and the ability to sort of dispose of vehicles quickly and probably cheaper than other subprime players could potentially allow you to have a better offer in the marketplace.

Matt, we dispose of a lot of our current repos through our auction channel, so I think it's just another thing that we do that makes it easy for us to run a finance business or easier to run a finance business, because we don't have to rely heavily on outside parties to dispose of repos. So it's a factor, but I don't think it's a big one.

Yes. And, Matt, I'd also point out, we're not -- in the event that we go through this test and we determine it's the right thing to do for us, our goals are not going to be to outcompete other people in the subprime space. We're looking at it as potential profit and risk mitigation in our business.

Okay, great. And then on the warranty business, I'm curious just why the cancel rate has ticked higher. It seems like at $10 a month on a $200 or $300 payment, it's not something that you would really pay a lot of attention to.

Yes, Matt, we're not going to be able to go into any granularity on what is driving the cancel rate, but we have observed a higher rate of cancellations on the ESP we've originated over recent years. And as a reminder, that product is cancelable anytime during its life, and we have to return a pro rata amount in the event the customer cancels it. They could cancel it at will. They could be canceling it because they're buying a new car, because they got repo-ed, for any number of reasons. So as we sell that product, we book a reserve, and it's an estimate, just like it would be on CAF loan losses, that we have to tune on a periodic basis. If you remember, we had a $0.01 adjustment in the fourth quarter. The portfolio has gotten a lot bigger. We've actually sold over $700 million of the product since 2010. And we've seen the movement, so it's appropriate to adjust the reserve amount here.

I guess, do you think that's -- should we view that as more of a macro consumer issue? Or is it -- does it have something to do with, you think, with your offer?

Like I said, we can't go to that kind of granularity. What I would view it as, from your perspective, is a very slight worsening in the profitability of ESP because we have a different view on what's going to be returned on a go-forward basis. And if we see different behavior going forward, we'll change it again. But it amounts to a small adjustment in the go-forward reserve, and a minimal, maybe 1% less profitable on the ESP that we sell. So it's a few dollars a car.

And then just lastly, the stats on the web visits and the mobile usage are pretty strong. What does that mean for your business model? I mean, are those customers more likely to convert? Are they more likely to refer you to a friend? Have you kind of thought through what that could mean from a financial standpoint, longer-term?

Yes, it's a little hard to say, but I just clearly think this is where consumers are going. And they're spending a lot more time on the web prior to coming to the store, and we've seen that trend for several years. So it's not surprising that the growth of web traffic dramatically exceeds the growth of foot traffic. I think people are showing up more prepared. So I think it has lots of ramifications for the business, but it's really difficult to figure out what that is right now. But in general, I think it's positive. I think we have a great app, and I think we have a great website and a great search engine. And I think we're providing tools that customers want so that they're better prepared to make a decision.

Your next question is from John Murray (sic) [Murphy].

It's John Murphy. Just a follow-up on the CAF question here on the subprime side. What is the motivating factor there? Is that really just to grow CAF earnings, which is a good thing, or is that a response to this pullback in credit availability you're seeing from your subprime partners? And also, what was the highest that subprime was as a percentage of sales at the peak?

Yes, John. As I mentioned, we've worked on this for over a year, so it's not a reactionary move at all. Customers with challenged credit have become a meaningful part of our overall business, and they're a meaningful part of the used car market, so we feel like we owe it to ourselves to get smarter about this space. Whether we're in it or not with our partners, we need to understand it a little better. We considered a number of options, including a joint venture or something like that, but we determined this is the best way to learn. So I hope -- that's probably the best way to address [indiscernible].

Yes, I mean, John, when this was 3% or 4% of our business, it wasn't as big of a discussion point. But when it's 15% plus, we have to look at what's in the overall best interest of the company. So this is just a step in that direction to really learn more about the segment.

It seems like it makes a lot of sense. But what was the peak percentage, is it 18% kind of...?

21%.

It's 21%.

21%, okay, great. Then just a second question on vehicle pricing, up $125. Tom, was that just -- is that just a function of mix? Because there's always this expectation -- or there's this expectation in the market that used vehicle pricing might come down, but that doesn't seem to be happening at all.

John, I missed the beginning part of that. What did you say, used vehicle pricing?

Up $125.

Your revenue per unit was up $125 year-over-year, and it's a pretty strong number. I'm just curious what you're seeing for used vehicle pricing. Is that mix or -- I mean, there's an expectation that used vehicle pricing will come down, and that's just not happening.

Yes, I've never seen it come down other than in the recession, when it dropped a couple thousand dollars in less than 6 months. So I mean, we don't view $100 as a very big number on an over $19,000 average retail, so we didn't really spend any time looking or talking about that.

Okay. And then just lastly, bench of general managers for new stores, where are you at on that process?

We feel like we're in really good shape. Our opening plan, as we've announced, 10 to 15 stores over the next 3 years. This is the first of those 3. As a percentage of the base of stores that we have, it's actually a smaller percentage than what we've grown in the past. And I think we've done a really nice job over the last few years of building up our pipeline, both in terms of real estate availability, people availability, getting folks trained and ready to go and having people prepped to move to new markets. So I feel like we're in really good shape.

And your next question is from Matt Fassler.

I'd like to dig a little bit deeper into subprime as well, and just get a better understanding from you as to sort of the operational differences in managing a subprime business from running CAF as it exists today. And also, if you could give us some sense as to the sources of funding as you launch your pilot? And then, if it's successful, whether you would anticipate funding subprime through the ABS market or through some other vehicle.

Sure, Matt. So I'm not going to go into a lot of detail about what we're doing as far as managing services. But our game plan is to randomly route a small percentage of customers to CAF and run them through the scoring model that we've developed and provide great customer-friendly service. One thing I think is important to remember, we've got a great team in Atlanta that runs a great finance business, and we already serve a pretty wide spectrum of credit. Some customers need more attention than the others as far as calls, et cetera. We're viewing this as an expansion, really, of what we already do. And like I said, without getting into details, we've built upon our account servicing techniques, and I think it will be appropriate -- it will be helpful for this space. But I think we'll also learn some things and have some benefit in the business we do today from it. As far as funding, at this point, we're just going to use cash on hand to fund the test, it's very small. To the extent we decide to go forward with this as a line of business or an initiative, we'll explore other options, and we would plan on trying to do something, to the extent it's available. Whether it's in the ABS market, conduits, that will be -- we'll determine that as we see fit.

And then just a quick follow-up, this just goes back to core used vehicle sales. Obviously, the supply of newer vehicles coming off lease seems to be starting to gush, and it's plentiful relative to where it had been in recent years. Are you seeing that flow through to your buying opportunities? What's your sense of whether that's actually materializing the way the numbers say it is and what that does for your sourcing and volume potential going forward?

Yes, it's -- I think the gush you're talking about in leasing is leased percentage of new cars sold, and that's really a lot more recent. So those cars, really, have not started coming back in any big numbers. So as we've talked about last quarter, that our mix shift between 0- to 4-year-old and 5- to 10-year-old really hasn't moved very much. But clearly, if the SAAR stays -- continues to move up in the 16 million plus range and the percentage of leased vehicles is higher, then, ultimately, those cars will come back at the auction. It really hasn't started to happen yet, but we expect it to, and I think that's good for us.

Your next question is from Sharon Zackfia. Sharon Zackfia - William Blair & Company L.L.C., Research Division: So a couple of questions on CarMax Auto Finance, and then I had a question for Tom as well. So on the tightening of credit, it didn't look like, based on the percent of subprime sales in the quarter, that it had any kind of noticeable impact. So has it been more of an impact as you've gone into the fourth quarter, or was it near the very end of the third quarter?

Yes, I think, as we mentioned in the release, late in the quarter we saw a change of behavior. And every call, last 3 -- 2 or 3 calls, I've been asked whether we see any reason to think that our partners' behavior may be different than it has been. And this time, we do. So that's why we're talking about it. Sharon Zackfia - William Blair & Company L.L.C., Research Division: Okay. And I mean, are the loans you're going to generate through CAF, are they expected to be a substitutive of the loans that you would have generated through the third-party lenders, so kind of cannibalizing the third-party? Or is it a significantly different scorecard than your third-party lenders use, so it would be additive, if I'm making sense there?

It's our own scorecard, but we plan on routing to CAF just like we route to our other third-party lenders. So it would eat a little bit into what they're doing.

We're not going into this, Sharon, thinking we're going to add a bunch of incremental sales because our third-party providers are missing the boat in this arena. This just becomes such a big percentage, we don't want to have kind of so little visibility into this area, particularly one that's so costly for us.

Right. It's important to remember that these sales through these partners are about 1/3 as profitable as something that goes through the -- either the CAF or one of our other channels. So there's a big economic [indiscernible] there between...

Yes, there's quite a lot of incentive there. Sharon Zackfia - William Blair & Company L.L.C., Research Division: And I assume, if this is a good pilot program and you expand it further sometime in the future, you wouldn't indefinitely fund this through your own balance sheet, right?

No. As I said to Matt, I think we'd look at alternatives to fund this. And as I also mentioned, we're looking at this as a play for profit, potentially, and risk mitigation. And at this point, if this test makes sense, we'd envision us being one of several partners in that space, not trying to grab all of the share. Sharon Zackfia - William Blair & Company L.L.C., Research Division: Okay. And then just a quick question on marketing. I mean the year-to-date marketing spend is pretty flat, which, I think, is relatively unusual. And I was hoping somebody could speak to whether that's just more efficiencies you're getting in the marketing team or in the buys that you have, if your impressions are actually up year-over-year or if those are flattish? Just any indication of what's going on with marketing because I'm just surprised that it's not up more.

Well, we have pretty good comps for the first 9 months of the year. So that, hopefully, is going to drive some leverage. And we do have some timing. We'll spend a little more on the second half of the year. We are planning along running a Super Bowl ad this year. That's -- it's a few million dollars. It's not dramatic, but it's -- there will be a little bit of timing into the fourth quarter. But we expect the business to leverage in a number of different areas. And if we're running -- we were 16, 17 and 10 in comps for the first 3 quarters, so... Sharon Zackfia - William Blair & Company L.L.C., Research Division: It was actually flat in dollars, though. I mean, I was just looking at it in dollars, not as a percentage of sales.

Yes, and we think of it as dollars per car sold. So that's -- when I say leverage, I'm talking about in dollars per car sold. Sharon Zackfia - William Blair & Company L.L.C., Research Division: Okay. You're going to force me to watch the Super Bowl, that's the master plan.

Yes.

Your next question is from Brian Nagel from Oppenheimer. Brian W. Nagel - Oppenheimer & Co. Inc., Research Division: So I, too, want to ask a question about subprime. If -- what can you -- is there any way to quantify the impact to your -- upon your sales of these actions the lenders have taken here recently? I mean, in other words, how should we think about how much that is, either in the current quarter or going forward? How much could this weigh upon your used car sales?

Yes. I mean, we aren't very good at predicting the future for any part of our business, which is why we don't give guidance. And I think it's just too early to tell. We don't know where they're going to land as far as where they dial in at how much they're going to tighten, so we can't give any guidance going forward. But we -- like I said, it is a -- we have observed a change. We get asked this question every quarter, and we thought it was important to let you know.

And there's way too many other variables.

Yes.

When you think like the question Matt asked about supply and customer traffic and the growth in our web traffic. There's just a lot of things, a lot of moving parts there, so it's very difficult to predict. Brian W. Nagel - Oppenheimer & Co. Inc., Research Division: Got it. And the follow-up question, someone previously asked about the performance of your new stores. Just wondered if you could clarify it. And just given that you opened some more new stores, how would you characterize overall the performance of the stores?

Yes, as a group, with all of our new -- the new stores are at or above our expectations. So we're very pleased with how openings have been going. And yes, it's a pretty diverse set of stores, too, spread out across the country, and that will be the case going forward as well. So just what we're really focused on is continuing to improve our business model each and every day and make sure that when we open a store, we have an experienced CarMax team in place, so when a customer walks in the door, they get a great experience. And we've been very pleased with how that's gone so far. Brian W. Nagel - Oppenheimer & Co. Inc., Research Division: And one more follow-up, if I could, just going back on subprime. Again, so I think someone else asked, is there -- could you -- is there any way you could tell the reason behind the change in behavior of your partners? The question I have, have you seen anything that suggests that the performance of these subprime loans is actually deteriorating?

Well, we don't participate in this space, so I can't really comment on that. But I think I would hypothesize that they're -- they have not been as happy with what they've seen -- that they've originated over the last 2 years, and they're adjusting their portfolio accordingly.

Look, a number of times over the last several quarters we've said that our subprime providers have gotten more and more comfortable and more and more aggressive with the CarMax origination channel. It just makes sense that, at some point, you'd bump up against some performance metric and probably have to scale back a little bit. But again, we don't run their business. They run their own business. We're happy to have them as third-party providers, and -- but we're going to expect some movement in the way they lend over time.

Yes. And as we mentioned, that profitability difference is significant, so we're actually happy to trade 3 of those for 1 CAF sale anytime you tell us and any time we could take that trade.

That's why it's another difficult thing to really figure out because, I mentioned all those other moving parts earlier, it doesn't take a lot of non-tier 3 sales to make up for 1 point or 2 of loss.

Your next question is from Craig Kennison from Robert W. Baird. Craig R. Kennison - Robert W. Baird & Co. Incorporated, Research Division: I'll start with a quick one on subprime and then move in a different direction. I assume that this has no impact on the fee that you will pay to your origination partners.

No, it doesn't. As Tom said, we don't influence their behavior. The one way that we could theoretically do it is by paying them more or less, but I think you've seen when we've announced our tests, we've elected to leave that as it is and figure out the business a little bit better ourselves. Craig R. Kennison - Robert W. Baird & Co. Incorporated, Research Division: That helps. And then just changing direction here, on the real estate side, Tom, I know part of the strong return on capital that you generate is because you've got a good real estate team that buys real estate at the right price. What are you seeing in that market today? And do you feel like you can still get the kind of returns as with those rates -- or those costs go up?

We feel pretty good. We've not committed to a single piece of real estate that we didn't think we could make a very good return on. So are there things that are -- coming out of the recession, were there things that were less expensive than they were before? Sure. And now, going into places like -- look -- it's starting to look at Portland, Seattle, San Francisco, Boston, Philly, are they places that are more expensive? Absolutely, but we expect to sell more cars there. And we've also done a nice job over the last few years of improving our profitability. So in terms of availability and our ability to make a great return, it's been no problem at all. Craig R. Kennison - Robert W. Baird & Co. Incorporated, Research Division: And with respect to reconditioning costs, obviously, you've moved the needle there over time. Is there still any room at all to reduce costs there?

Yes, we do still think there's some room. I think it's -- as we've said before, the next 50 or 100 will be a lot more difficult than the first 250 or so. But it's a constant effort for our teams to try to figure out how to get more consistent and how to eliminate waste from the system and, at the same time, not compromise our quality. So I think our store teams have done a fantastic job to this point, and I expect to still be able to make progress going forward.

And your next question is from Scot Ciccarelli from RBC Capital Markets.

It will be remiss of me not to ask at least one subprime question here.

We might have to valid check [ph] you and just say, "Look, we've already answered that question 7 times."

Okay. And I've heard that from you before, actually. So I guess my question is, where are you comfortable with subprime penetration levels? I mean, historically, you guys have kind of talked about -- Tom, you've kind of talked about 20%-ish kind of level. But now you're experimenting it with -- experimenting yourselves with subprime. Like has the thought process in terms of how much subprime is going to be as a total of your business started to change?

Scot, I don't think it has at all. We don't know the right number, but one thing, I think, we need to be clear about is we're not intending to drive subprime penetration at our business up by getting into this initiative here. As I said, we're looking at potentially being one of several partners. We'll have our own set of credit parameters and our own origination model, which we do not expect to be more aggressive than our partners. So I don't think we're giving any indication that we think we're moving -- trying to move the needle on how much subprime is in the business.

Yes, that's correct. That's not our goal at all in doing this. We don't expect this to have, really, any impact whatsoever on the percentage of sales represented by subprime.

Yes, we're looking at whether it makes sense to take a share of the profit and to try to mitigate risk.

We've been getting asked for years about the amount of money we pay to get one of these deals done and wouldn't it make sense to think about it differently, and that's what we're trying to do.

Understood. So it's really a substitution effect here. But historically, you've also talked about you're kind of comfortable with 20% penetration levels of subprime. Does that thought process change if the new model is, let's call it, significantly more profitable than what the old subprime model had been?

No, I don't think it does. As I said, we're not looking to expand the amount of subprime in our channel. And when you look at the amount of subprime business that we do, it's not out of line with the marketplace overall. We've got -- we've looked at external data to that point, so we're not -- I think subprime is going to be what it is based on customer flow through the door and who's applying and what's going on in the economy. And we'll run a business that makes -- if we -- if this test is successful, we'll run a business that makes economic sense. But I don't think we have -- I feel comfortable that we're not out of line with the mix of subprime in the auto space, in the used auto space, and we're not trying to drive it up.

And, Scot, we've never really put a number on that. What I've always said about this segment of business is each incremental -- each individual customer, first of all, is getting a great deal at CarMax because we're not changing the price because they happen to be in this segment. Second of all, we've been subsidizing the deal by paying a discount to get them done. And we deliver very high-quality products. So I think if you're in this credit space, we're a terrific place to buy a very high-quality car and get a great experience that maybe you can't get elsewhere. And also, since these lenders don't see the loan until all of the other providers have declined, we've been sure that these customers and these sales are 100% incremental. So despite the fact that they're lesser in terms of profitability, it's 100% incremental profit. So on an individual customer basis, we feel really, really good about this offering. And as Tom said, when you look at it as a percentage of our total sales, we're going to sell 0.5 million cars this year. You would expect us to be somewhat representative of what kind of credit mix looks like in the U.S., and that's kind of where we are right now.

Okay, understood. I will ask about the test that you're doing, though. Like some of your -- some of the loans within your typical CAF portfolio are already subprime if you just kind of look at what FICO scores are and APRs and the securitizations. Why the call out about the test? Because you don't want the blended FICO score or credit quality of existing securitizations to change?

Well, if you think of -- you're absolutely right, Scot, that we do serve a wide spectrum of credit, as I mentioned before, and we're looking at this as expansion. But the way credit routes in our business is it goes to CAF, it goes to our Tier 2 lenders, it goes to our Tier 3 lenders. We want our customers to have an opportunity to get all those various offers and get the best financing for them. And this is just -- in this particular space are customers that we have not served before, and...

But I would say that you're right, in that some of the stuff we originate ends up performing like this, which gives us more confidence in our ability to manage it. Because we've been managing receivables, although it's a small percentage, it's still a big enough chunk that we feel like we have experience to do this well. And part of calling it out is to separate it out and look at it as a separate investment and see if we can make a return on it. But if we just originated this different set of business and put it into the securitization, we wouldn't get as good -- as favorable terms as we're currently getting. And once we get this all -- a test in place and we decide whether or not we want to go forward, we'll figure out how to fund it then.

Okay, that's all very helpful. And then just one more question regarding wholesale. Looks like wholesale units were a bit sluggish. I know you guys don't break out comps for the business, but given the store growth, it would suggest that the comp will be in negative territory, which just seems odd given kind of the strength in the rest of your business. Any thoughts there, Tom?

No, I still think -- I've always thought that wholesale and retail, over time, will grow around the same. And you've been following us for a long time, you've seen huge swings that don't necessarily match up. If you go back to 2001, I know that's a long term to look at it, our wholesale growth and our retail comps have been about the same, but -- and I expect that over time. So 4% this quarter compared to 15% sales growth is a little bit off, but I think, over time, they'll be more closely aligned.

But nothing that you noticed in terms of why the wholesale unit growth was slower this quarter, maybe buying fewer vehicles or something?

No, I mean, clearly, we bought more because we had growth, but not as many as what our sales growth would have predicted. But in terms of our buy rate, it was very strong, and we don't see any reason to worry about it.

And your next question is from Rod Lache from Deutsche Bank.

It's Dan Galves for Rod. I came on the call a little late, has anybody asked about subprime yet?

That was good.

So I got a couple of questions. The first one is related to the new stores that are kind of now in the comp base, would expect that they're growing faster as they ramp up to the corporate average. Do you see that as providing a meaningful positive to your same-store -- overall same-store comps yet? Or is that something still to come?

It's all just built into the comp expectation. And remember, too, that the base is so much bigger that even when you get a store at, say, its second year and it has a higher percentage comp growth, it's a very small number of units in the big scheme of things, so...

But as you end up with like 30, 40 stores in the next couple of years that are in year 2 and year 3, do you see that as being a potential positive to comps?

Yes, but as we've talked about a comp range in the past of something like 4% to 8%, we think that range is still a pretty good one. One other thing to remember is we're going to open a lot of stores back into existing markets. So sometimes, a satellite store can actually hurt comps in the other stores because there is some cannibalization. So it's a pretty complex model to figure out.

Okay, got it. Got it. And then the last one is the third-party finance fees, just rough math, if you divide that by the subprime volume in your business, you get to about $800 per unit. Is that the right way to think about it? Or are there -- is there third-party finance income in there as well? It's just looking -- go ahead.

We've said this before, in our subprime segment, today, we pay about $1,000 a car to the third-party provider.

And your next question is from Rick Nelson from Stephens.

I'd like to ask you about provision for loan loss. It looks like it picked up sequentially. If you could speak to delinquencies, what you're seeing there.

The provision for loan loss picked up dollar-wise year-over-year, but as a percent of receivables, it's flat. And if you're saying sequentially over the quarter, this is the time of year where we'll typically see, coming into the holiday season, delinquencies may pick up a little bit, but that's a normal seasonal trend. People are focused on other payments.

Got it. Also, I'm curious about the CFPB, they've put out a bulletin in March, how that's affecting your other third-party credit providers and if that's having any impact at CAF?

I don't -- to date, it's had no impact to CAF as we have not been visited yet. We have spoken with our third-party providers who have had interaction with the CFPB. I don't believe it's impacting any of our business. It's -- their focus is on a couple of things: one is making sure that people are treated equally and fairly; two, that the products aren't overly profitable. But as I said, as far as Wells Fargo or Cap One and the impact on the business they're doing with us, we haven't seen any.

And your next question is from Jamie Albertine from Stifel. James J. Albertine - Stifel, Nicolaus & Co., Inc., Research Division: I dialed in a little bit late. I just wanted to follow up, if I could, on the SG&A side of things. We've talked about it both with you guys at your Analyst Day and I've heard you referred to it on prior conference calls. But from an SG&A per unit perspective, can you just remind us kind of where we are today versus prior peak? And you've done a really good job of sort of highlighting the opportunities on items like reconditioning costs coming down, and it sounds like there's still room to go there. But maybe if you can talk sort of high level about your top 2 or 3 priorities as it relates to SG&A cost reductions and, therefore, leverage as you -- your growth story sort of plays out for the next 2 or 3 years.

Yes, I don't really know what the peak was, but we were down almost $100 per unit sold in the third quarter compared to last year. You referenced reconditioning costs. Remember, reconditioning costs are in cost of goods sold, not in SG&A, so we kind of try to think of those 2 things as 1 big bucket of opportunity. We will eclipse $1 billion of SG&A this year and probably, I don't know, $600 million or so of reconditioning. So it's a pretty big opportunity to try to find efficiencies. And in terms of our priorities, it's largely -- in the SG&A bucket, it's largely going to be around how do we operate our stores consistently, how do we staff better, how do we make sure we're properly staffed for peaks and for valleys in terms of traffic and sales. And it's a pretty seasonal business, it's a pretty labor-intensive business, so there's lots of opportunities to try and get more and more efficient in that regard. So that's kind where our efforts are focused there. We're really proud of all the progress we've made on the cost of goods sold side in terms of reconditioning. But I think with SG&A, there's probably some opportunity in there, and we're focused on going after it. James J. Albertine - Stifel, Nicolaus & Co., Inc., Research Division: I should have clarified that, and I apologize, and I appreciate you calling that out on the COGS side. Is it fair to say then that if you look at the 2 buckets, it's predominantly going to be more SG&A-driven from this point going forward, given that the bulk of reconditioning improvement you've already performed [ph]?

Not necessarily because the bulk -- remember, the bulk of SG&A is going to be rents and advertising. And some of the fixed overheads we can only make so much progress on. We have to have enough people in our stores to have them up and running and pay our bills. So I think when you look at the 2 buckets and you break it out into things that are somewhat less manageable, let's say, then there's opportunity in both.

And your final question is from Bill Armstrong from CL King & Associates. William R. Armstrong - CL King & Associates, Inc., Research Division: Tom and Tom, this is the first we've heard of any hiccups in the overall subprime model lending market. And you mentioned earlier that you do not believe that this is CarMax-specific. What -- I was wondering if you could maybe elaborate on what you're hearing or seeing in the market that leads you to believe that this is beyond CarMax?

No, I think -- we've had conversations with our partners. And as I've mentioned, they have CarMax as a part of their overall portfolio. And they fine tune that portfolio and their origination standards based on what they're experiencing. And I think that is the result of us, other business they're doing. And all I can say is that they have indicated that this is not a CarMax-specific issue. William R. Armstrong - CL King & Associates, Inc., Research Division: If the subprime consumer is starting to get a little overextended with debt, particularly auto-related debt, how does that impact your view on this test that you're now doing? And it sounds like it might be -- you might be extending that at just the wrong time. How do you look at that?

If you believe the market has been overheated over the last couple years, then the last couple years would have been the wrong time to enter this. We're not trying to time the market. We're not trying to figure out what's the best time to enter it. This is a longer-term initiative. The reason we're going now is because we're ready to go now. We've been working on it for over a year. And we're in a position where we believe we can start originating and servicing sometime this quarter. And remember, this is a very small test for us. This is smaller than other tests we run at CAF as far as pricing and in terms. And at the end of the day, while we are going into it expecting to make some money, it's not material to our overall business or even to CAF's overall portfolio. William R. Armstrong - CL King & Associates, Inc., Research Division: Understood. Okay. And then just one quick housekeeping question. The increase in allowance on ESP returns, what was the dollar amount on a pretax basis?

It's about -- it's $0.02. William R. Armstrong - CL King & Associates, Inc., Research Division: Right. What is the dollar amount on a pretax basis?

Yes, we don't disclose -- we haven't disclosed dollar amounts on our estimate adjustments.

And I'm showing no further questions at this time.

Okay. Thanks, everyone. Happy holidays. Thanks for all your support and continued interest. And of course, thanks to all of our associates for their dedication and their hard work and what they do every day to make CarMax what it is today. We'll talk to you again next quarter. Thanks.

And that does conclude today's conference. You may now disconnect.